Benchmarks by Michael Larabel for a future article.
EPYC 8534P Processor: AMD EPYC 8534P 64-Core @ 2.30GHz (64 Cores / 128 Threads), Motherboard: AMD Cinnabar (RCB1009C BIOS), Chipset: AMD Device 14a4, Memory: 6 x 32GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92 + 2000GB Corsair MP700 PRO, Graphics: ASPEED, Network: 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00213Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
EPYC 8534PN Changed Processor to AMD EPYC 8534PN 64-Core @ 2.00GHz (64 Cores / 128 Threads) .
Changed Disk to 2 x 1920GB KIOXIA KCD8XPUG1T92 .
EPYC 9654 Processor: AMD EPYC 9654 96-Core @ 2.40GHz (96 Cores / 192 Threads)Motherboard: AMD Titanite_4G (RTI1007B BIOS)Chipset: AMD Device 14a4, Memory: 12 x 64GB DDR5-4800MT/s Samsung M321R8GA0BB0-CQKEGDisk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
EPYC 9654 2P Changed Processor to 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads) .
Changed Memory to 1520GB .
EPYC 9684X Changed Processor to AMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads) .
Changed Memory to 12 x 64GB DDR5-4800MT/s Samsung M321R8GA0BB0-CQKEG .
EPYC 9684X 2P Changed Processor to 2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads) .
Changed Memory to 1520GB .
EPYC 9754 Changed Processor to AMD EPYC 9754 128-Core @ 2.25GHz (128 Cores / 256 Threads) .
Changed Memory to 12 x 64GB DDR5-4800MT/s Samsung M321R8GA0BB0-CQKEG .
Processor Change: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116
EPYC 9754 2P Changed Processor to 2 x AMD EPYC 9754 128-Core @ 2.25GHz (256 Cores / 512 Threads) .
Changed Memory to 1520GB .
Xeon Max 9468 Processor: Intel Xeon Max 9468 @ 3.50GHz (48 Cores / 96 Threads)Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS)Chipset: Intel Device 1bceMemory: 576GBDisk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Xeon Max 9468 2P Changed Processor to 2 x Intel Xeon Max 9468 @ 3.50GHz (96 Cores / 192 Threads) .
Changed Memory to 1136GB .
Xeon Max 9480 Changed Processor to Intel Xeon Max 9480 @ 3.50GHz (56 Cores / 112 Threads) .
Changed Memory to 576GB .
Xeon Max 9480 2P Changed Processor to 2 x Intel Xeon Max 9480 @ 3.50GHz (112 Cores / 224 Threads) .
Changed Memory to 1136GB .
Xeon Platinum 8490H Changed Processor to Intel Xeon Platinum 8490H @ 3.50GHz (60 Cores / 120 Threads) .
Changed Memory to 512GB .
Processor Change: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b000590
Xeon Platinum 8490H 2P Changed Processor to 2 x Intel Xeon Platinum 8490H @ 3.50GHz (120 Cores / 240 Threads) .
Changed Memory to 1008GB .
Xeon Platinum 8592+ Changed Processor to INTEL XEON PLATINUM 8592+ @ 3.90GHz (64 Cores / 128 Threads) .
Changed Memory to 512GB .
Processor Change: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000200
Xeon Platinum 8592+ 2P Changed Processor to 2 x INTEL XEON PLATINUM 8592+ @ 3.90GHz (128 Cores / 256 Threads) .
Changed Memory to 1008GB .
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Weld Porosity Detection FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 10K 20K 30K 40K 50K SE +/- 0.41, N = 3 SE +/- 1.69, N = 3 SE +/- 4.46, N = 3 SE +/- 147.69, N = 3 SE +/- 11.71, N = 3 SE +/- 110.20, N = 3 SE +/- 5.50, N = 3 SE +/- 11.27, N = 3 SE +/- 234.43, N = 4 SE +/- 80.58, N = 3 SE +/- 21.27, N = 3 SE +/- 233.95, N = 3 SE +/- 2.71, N = 3 SE +/- 53.98, N = 3 SE +/- 47.41, N = 3 SE +/- 39.08, N = 3 5366.79 4830.01 9356.71 18666.16 9997.89 20550.93 11508.19 22499.03 19183.83 37710.60 20910.53 40032.81 15943.17 47654.95 26943.42 48122.16 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Degridding EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 30K 60K 90K 120K 150K SE +/- 44.43, N = 3 SE +/- 69.57, N = 3 SE +/- 189.10, N = 3 SE +/- 882.91, N = 4 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 598.81, N = 12 SE +/- 1077.92, N = 14 SE +/- 0.00, N = 3 SE +/- 143.60, N = 3 SE +/- 76.47, N = 3 SE +/- 0.00, N = 3 SE +/- 93.91, N = 3 SE +/- 378.22, N = 4 SE +/- 153.57, N = 3 SE +/- 480.95, N = 3 16704.4 15903.4 42172.4 81025.8 64260.2 123480.0 28476.6 52639.5 18522.0 36900.5 20561.4 40368.6 17891.6 34614.1 31177.2 59139.2 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Result
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 30K 60K 90K 120K 150K SE +/- 86.77, N = 3 SE +/- 0.00, N = 3 SE +/- 1853.72, N = 3 SE +/- 2386.93, N = 4 SE +/- 581.17, N = 3 SE +/- 19579.11, N = 3 SE +/- 685.49, N = 12 SE +/- 1180.69, N = 14 SE +/- 99.83, N = 3 SE +/- 194.30, N = 3 SE +/- 0.00, N = 3 SE +/- 802.88, N = 3 SE +/- 76.67, N = 3 SE +/- 412.68, N = 4 SE +/- 219.20, N = 3 SE +/- 1080.03, N = 3 16573.2 16399.7 46233.7 86757.7 74389.0 121976.9 28854.3 54066.0 21766.5 42939.2 23250.2 45195.1 21314.2 41773.0 37265.9 72004.0 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp Mpix/sec Per Watt
OpenBenchmarking.org Mpix/sec Per Watt, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 130.96 142.12 188.14 167.00 247.30 199.37 134.70 109.74 69.96 67.77 75.14 70.87 67.85 66.72 125.14 128.04 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better ASKAP 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.03 / Avg: 126.56 / Max: 161.28 Min: 16.5 / Avg: 115.4 / Max: 136.82 Min: 29.47 / Avg: 245.74 / Max: 336.85 Min: 57.21 / Avg: 519.5 / Max: 681.6 Min: 31.29 / Avg: 300.8 / Max: 409.34 Min: 64.2 / Avg: 611.81 / Max: 819.12 Min: 20.61 / Avg: 214.21 / Max: 327.89 Min: 37.58 / Avg: 492.69 / Max: 670.56 Min: 74.39 / Avg: 311.12 / Max: 354.1 Min: 196.26 / Avg: 633.57 / Max: 720.17 Min: 73.2 / Avg: 309.44 / Max: 360.17 Min: 196.89 / Avg: 637.73 / Max: 729.76 Min: 57.81 / Avg: 314.14 / Max: 357.36 Min: 163.2 / Avg: 626.07 / Max: 724.87 Min: 24.59 / Avg: 297.8 / Max: 352.67 Min: 74.42 / Avg: 562.35 / Max: 727.39
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 80K 160K 240K 320K 400K SE +/- 23.20, N = 3 SE +/- 14.31, N = 3 SE +/- 72.03, N = 4 SE +/- 1929.89, N = 5 SE +/- 1178.78, N = 6 SE +/- 1666.53, N = 6 SE +/- 126.50, N = 4 SE +/- 2064.71, N = 6 SE +/- 67.02, N = 3 SE +/- 679.14, N = 4 SE +/- 9.58, N = 3 SE +/- 420.81, N = 4 SE +/- 150.79, N = 3 SE +/- 560.20, N = 4 SE +/- 31.76, N = 4 SE +/- 1528.09, N = 5 52745.64 52684.93 125191.11 256584.67 225311.82 390520.93 127519.37 214017.18 58815.36 135939.25 69776.41 141954.15 71245.26 147311.14 109416.42 228184.68 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6 Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200 400 600 800 1000 393.05 441.50 547.32 649.64 882.53 916.62 590.51 567.85 189.97 243.32 229.75 251.86 240.55 275.88 411.62 515.52 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.23 / Avg: 134.2 / Max: 159.66 Min: 17.36 / Avg: 119.33 / Max: 138.69 Min: 26 / Avg: 228.74 / Max: 319.02 Min: 56.6 / Avg: 394.96 / Max: 670.11 Min: 32.04 / Avg: 255.3 / Max: 409.27 Min: 63.25 / Avg: 426.04 / Max: 818.12 Min: 22.12 / Avg: 215.95 / Max: 323.68 Min: 41.44 / Avg: 376.89 / Max: 668.41 Min: 73.26 / Avg: 309.61 / Max: 365.04 Min: 187.62 / Avg: 558.68 / Max: 751.91 Min: 74.98 / Avg: 303.7 / Max: 376.9 Min: 183.64 / Avg: 563.63 / Max: 754.82 Min: 60.29 / Avg: 296.18 / Max: 370.22 Min: 158.17 / Avg: 533.96 / Max: 753.09 Min: 26.78 / Avg: 265.82 / Max: 376.59 Min: 68.95 / Avg: 442.63 / Max: 754.48
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.3 Algorithm: SHA256 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70000M 140000M 210000M 280000M 350000M SE +/- 10642001.84, N = 3 SE +/- 64797308.82, N = 3 SE +/- 38601729.93, N = 3 SE +/- 210407867.05, N = 3 SE +/- 146384349.91, N = 3 SE +/- 160246527.97, N = 3 SE +/- 34182873.86, N = 3 SE +/- 80639802.67, N = 3 SE +/- 27466135.54, N = 3 SE +/- 126029257.64, N = 3 SE +/- 33334799.97, N = 3 SE +/- 70647143.49, N = 3 SE +/- 159515825.53, N = 3 SE +/- 66583540.43, N = 3 SE +/- 23323764.84, N = 3 SE +/- 29142783.86, N = 3 74584721007 67806730387 126475137690 251645629887 138409588983 276267766403 161269162280 320258686237 44439185187 88635753413 51840436763 103382860113 61806994523 119106802107 66573957920 131976050120 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.3 Algorithm: SHA256 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110M 220M 330M 440M 550M 469088919.27 511227929.81 386445411.79 378210579.19 345260515.93 344443539.20 481947170.76 471857277.56 157500377.52 149911179.45 167457577.18 160789835.78 189585676.93 179424218.79 195213927.96 193109819.08 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.27 / Avg: 159 / Max: 170.75 Min: 17.53 / Avg: 132.64 / Max: 140.44 Min: 29.19 / Avg: 327.28 / Max: 361.5 Min: 59.61 / Avg: 665.36 / Max: 734.49 Min: 34.63 / Avg: 400.88 / Max: 411.05 Min: 69.9 / Avg: 802.07 / Max: 822.12 Min: 22.61 / Avg: 334.62 / Max: 353.13 Min: 42.79 / Avg: 678.72 / Max: 712.84 Min: 75.75 / Avg: 282.15 / Max: 323.37 Min: 182.49 / Avg: 591.26 / Max: 697.86 Min: 82.11 / Avg: 309.57 / Max: 363.52 Min: 186.31 / Avg: 642.97 / Max: 734.25 Min: 57.24 / Avg: 326.01 / Max: 380.05 Min: 163.14 / Avg: 663.83 / Max: 754.31 Min: 32.34 / Avg: 341.03 / Max: 377.49 Min: 71.86 / Avg: 683.42 / Max: 754.43
Pennant Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: sedovbig EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.024479, N = 4 SE +/- 0.029749, N = 4 SE +/- 0.052975, N = 7 SE +/- 0.040626, N = 15 SE +/- 0.080453, N = 15 SE +/- 0.041359, N = 15 SE +/- 0.074320, N = 15 SE +/- 0.037087, N = 15 SE +/- 0.001847, N = 3 SE +/- 0.003237, N = 4 SE +/- 0.013353, N = 3 SE +/- 0.003253, N = 4 SE +/- 0.002987, N = 3 SE +/- 0.016400, N = 4 SE +/- 0.002400, N = 3 SE +/- 0.018471, N = 4 11.369640 12.151320 5.916458 3.408337 5.460660 3.081108 5.147658 2.940456 21.021120 13.211250 19.882460 12.106900 16.030800 11.284380 16.969100 9.964526 1. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Pennant 1.0.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.05 / Avg: 123.57 / Max: 166.99 Min: 16.13 / Avg: 110.49 / Max: 145.79 Min: 26.98 / Avg: 200.64 / Max: 329.21 Min: 55.22 / Avg: 351.84 / Max: 674.58 Min: 30.95 / Avg: 242.71 / Max: 409.53 Min: 64.82 / Avg: 404.01 / Max: 819.66 Min: 20.25 / Avg: 182.99 / Max: 330.83 Min: 41.72 / Avg: 310.27 / Max: 663.85 Min: 72.63 / Avg: 303.18 / Max: 377.15 Min: 189.6 / Avg: 573.95 / Max: 754.13 Min: 78.91 / Avg: 298.37 / Max: 377.04 Min: 206.49 / Avg: 569.16 / Max: 751.77 Min: 57.56 / Avg: 287.94 / Max: 376.04 Min: 148.41 / Avg: 535.71 / Max: 753.56 Min: 26.41 / Avg: 279.92 / Max: 376.36 Min: 69.95 / Avg: 489.66 / Max: 753.78
Xmrig Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org H/s, More Is Better Xmrig 6.21 Variant: GhostRider - Hash Count: 1M EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 7K 14K 21K 28K 35K SE +/- 30.53, N = 3 SE +/- 41.94, N = 3 SE +/- 117.32, N = 3 SE +/- 290.50, N = 3 SE +/- 10.36, N = 3 SE +/- 95.10, N = 3 SE +/- 61.27, N = 3 SE +/- 46.94, N = 3 SE +/- 12.93, N = 3 SE +/- 14.02, N = 3 SE +/- 14.78, N = 3 SE +/- 7.44, N = 3 SE +/- 23.16, N = 3 SE +/- 4.03, N = 3 SE +/- 44.68, N = 3 SE +/- 64.69, N = 3 4660.0 4564.3 14207.7 26702.7 15786.6 31157.8 9320.5 17873.1 5446.8 10975.8 6122.3 12115.0 5808.8 11263.2 8378.3 16324.7 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc H/s Per Watt
OpenBenchmarking.org H/s Per Watt, More Is Better Xmrig 6.21 Variant: GhostRider - Hash Count: 1M EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 11 22 33 44 55 33.47 36.24 46.69 45.05 42.00 44.08 40.17 36.17 16.64 16.29 17.93 18.00 17.11 16.83 25.02 25.29 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Xmrig 6.21 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.62 / Avg: 139.24 / Max: 143.75 Min: 15.67 / Avg: 125.96 / Max: 131.33 Min: 30.21 / Avg: 304.31 / Max: 328.13 Min: 54.62 / Avg: 592.72 / Max: 673.62 Min: 36 / Avg: 375.84 / Max: 410.13 Min: 60.4 / Avg: 706.81 / Max: 819.97 Min: 21.91 / Avg: 232.05 / Max: 250.74 Min: 38.53 / Avg: 494.18 / Max: 551.71 Min: 73.18 / Avg: 327.41 / Max: 335.76 Min: 201.16 / Avg: 673.63 / Max: 702.68 Min: 75.82 / Avg: 341.47 / Max: 356.09 Min: 188.92 / Avg: 672.99 / Max: 738.06 Min: 55.89 / Avg: 339.42 / Max: 349.66 Min: 159.48 / Avg: 669.3 / Max: 730.91 Min: 24.38 / Avg: 334.94 / Max: 357.54 Min: 73.19 / Avg: 645.48 / Max: 743.36
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.3 Algorithm: SHA512 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20000M 40000M 60000M 80000M 100000M SE +/- 15082903.45, N = 3 SE +/- 9990938.84, N = 3 SE +/- 6285194.15, N = 3 SE +/- 37307640.51, N = 3 SE +/- 55712209.05, N = 3 SE +/- 44023374.36, N = 3 SE +/- 5822493.86, N = 3 SE +/- 17716195.81, N = 3 SE +/- 500176.69, N = 3 SE +/- 3500318.33, N = 3 SE +/- 1250834.62, N = 3 SE +/- 14339684.98, N = 3 SE +/- 203213936.42, N = 3 SE +/- 481040738.21, N = 3 SE +/- 37654528.63, N = 3 SE +/- 14734422.48, N = 3 24378460820 22194470973 39627575497 78822257383 45865131913 91455971200 52746634113 104776633047 16199261277 32304222993 18893980783 37564522053 22219927723 42846254477 23886054770 46520760707 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.3 Algorithm: SHA512 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40M 80M 120M 160M 200M 154425638.15 165874523.26 128564080.98 125714359.16 114623223.18 114225793.10 157697655.07 154893248.59 54000562.67 51580447.52 57121711.23 55449110.37 65144677.30 62457308.50 69884736.57 67966151.09 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.44 / Avg: 157.87 / Max: 169.7 Min: 15.55 / Avg: 133.8 / Max: 143.63 Min: 28.99 / Avg: 308.23 / Max: 338.17 Min: 59.93 / Avg: 626.99 / Max: 687.56 Min: 34.64 / Avg: 400.14 / Max: 413.51 Min: 66.79 / Avg: 800.66 / Max: 822.16 Min: 21.49 / Avg: 334.48 / Max: 349.67 Min: 42.68 / Avg: 676.44 / Max: 705.8 Min: 71.29 / Avg: 299.98 / Max: 317.48 Min: 190.86 / Avg: 626.29 / Max: 729.82 Min: 74.41 / Avg: 330.77 / Max: 354.3 Min: 194.96 / Avg: 677.46 / Max: 722.73 Min: 56.54 / Avg: 341.09 / Max: 377.07 Min: 158.63 / Avg: 686.01 / Max: 753.85 Min: 23.35 / Avg: 341.79 / Max: 377.28 Min: 66.13 / Avg: 684.47 / Max: 754.35
7-Zip Compression This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300K 600K 900K 1200K 1500K SE +/- 120.28, N = 3 SE +/- 81.21, N = 3 SE +/- 111.53, N = 3 SE +/- 15794.26, N = 3 SE +/- 1085.99, N = 3 SE +/- 38431.24, N = 3 SE +/- 291.68, N = 3 SE +/- 14188.09, N = 3 SE +/- 916.22, N = 3 SE +/- 2598.93, N = 3 SE +/- 374.59, N = 3 SE +/- 6276.88, N = 3 SE +/- 61.74, N = 3 SE +/- 185.14, N = 3 SE +/- 119.21, N = 3 SE +/- 28320.74, N = 3 367989 338355 642626 1033277 716065 1116465 780871 1340226 207738 401707 243347 469131 287862 557054 338146 573410 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC MIPS Per Watt
OpenBenchmarking.org MIPS Per Watt, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 700 1400 2100 2800 3500 2768.54 2828.90 2434.84 2011.14 2281.23 1928.99 3140.68 2852.08 663.36 637.01 773.01 746.10 934.34 908.47 1119.86 986.76 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better 7-Zip Compression 22.01 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.72 / Avg: 132.92 / Max: 171.37 Min: 18.26 / Avg: 119.61 / Max: 153.19 Min: 27.34 / Avg: 263.93 / Max: 352.58 Min: 60.89 / Avg: 513.78 / Max: 716.51 Min: 33.61 / Avg: 313.89 / Max: 411.35 Min: 68.73 / Avg: 578.78 / Max: 823.64 Min: 22.44 / Avg: 248.63 / Max: 349.11 Min: 42.06 / Avg: 469.91 / Max: 704.89 Min: 70.41 / Avg: 313.16 / Max: 376.07 Min: 205.27 / Avg: 630.61 / Max: 754.38 Min: 74.78 / Avg: 314.81 / Max: 375.55 Min: 198.7 / Avg: 628.77 / Max: 756.25 Min: 60.14 / Avg: 308.09 / Max: 375.76 Min: 170.4 / Avg: 613.18 / Max: 756.05 Min: 34.7 / Avg: 301.96 / Max: 375.99 Min: 77.81 / Avg: 581.11 / Max: 756.45
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: Blowfish EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90K 180K 270K 360K 450K SE +/- 13.72, N = 3 SE +/- 20.54, N = 3 SE +/- 38.33, N = 3 SE +/- 3455.48, N = 3 SE +/- 98.33, N = 3 SE +/- 454.86, N = 3 SE +/- 1.33, N = 3 SE +/- 2397.15, N = 3 SE +/- 17.33, N = 3 SE +/- 116.69, N = 3 SE +/- 8.67, N = 3 SE +/- 169.28, N = 3 SE +/- 11.39, N = 3 SE +/- 348.45, N = 3 SE +/- 79.75, N = 3 SE +/- 649.08, N = 3 102198 93473 166541 316492 192574 374589 220188 415412 67893 134562 79152 156658 94513 183583 105831 205499 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2 Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 2023.03.14 Test: Blowfish EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200 400 600 800 1000 753.26 805.43 605.49 568.71 531.08 527.64 757.83 711.80 257.14 243.04 274.74 258.82 305.95 291.56 339.84 332.35 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better John The Ripper 2023.03.14 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.33 / Avg: 135.67 / Max: 154.48 Min: 15.84 / Avg: 101.56 / Max: 134.16 Min: 28.03 / Avg: 274.61 / Max: 310.69 Min: 54.52 / Avg: 556.51 / Max: 637.38 Min: 32.66 / Avg: 362.61 / Max: 410.64 Min: 67.44 / Avg: 709.94 / Max: 820.99 Min: 22.18 / Avg: 290.55 / Max: 334.66 Min: 41.23 / Avg: 584.92 / Max: 678.42 Min: 71.89 / Avg: 264.78 / Max: 291.9 Min: 183.06 / Avg: 553.66 / Max: 618.19 Min: 71.44 / Avg: 288.1 / Max: 318.09 Min: 196.84 / Avg: 607.6 / Max: 711.09 Min: 54.97 / Avg: 308.91 / Max: 345.11 Min: 168.16 / Avg: 627.31 / Max: 720.15 Min: 23.7 / Avg: 308.68 / Max: 376.19 Min: 71.27 / Avg: 618.12 / Max: 752.32
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: bcrypt EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90K 180K 270K 360K 450K SE +/- 22.28, N = 3 SE +/- 0.00, N = 3 SE +/- 20.00, N = 3 SE +/- 710.46, N = 3 SE +/- 154.46, N = 3 SE +/- 1315.55, N = 3 SE +/- 1.33, N = 3 SE +/- 4958.68, N = 4 SE +/- 8.84, N = 3 SE +/- 866.68, N = 3 SE +/- 8.67, N = 3 SE +/- 576.80, N = 3 SE +/- 10.67, N = 3 SE +/- 292.24, N = 3 SE +/- 46.28, N = 3 SE +/- 288.38, N = 3 102222 93472 166541 320527 192898 376268 220188 412723 67875 133866 79144 155959 94551 183603 105781 206778 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2 Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 2023.03.14 Test: bcrypt EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200 400 600 800 1000 749.17 920.41 606.47 573.89 531.63 523.96 759.60 705.60 256.35 241.56 273.70 256.68 305.82 292.69 342.68 334.53
OpenRadioss OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/ and https://github.com/OpenRadioss/ModelExchange/tree/main/Examples. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Chrysler Neon 1M EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 SE +/- 0.47, N = 3 SE +/- 0.50, N = 3 SE +/- 1.51, N = 8 SE +/- 0.71, N = 12 SE +/- 0.49, N = 3 SE +/- 0.46, N = 3 SE +/- 1.11, N = 3 SE +/- 1.00, N = 10 SE +/- 0.08, N = 3 SE +/- 0.13, N = 3 SE +/- 0.18, N = 3 SE +/- 0.14, N = 3 SE +/- 0.15, N = 3 SE +/- 0.11, N = 3 SE +/- 0.20, N = 3 SE +/- 0.14, N = 3 297.25 300.69 176.14 77.64 77.32 49.55 231.96 93.99 228.74 125.15 212.99 117.91 207.20 115.89 157.56 86.93 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenRadioss 2023.09.15 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.8 / Avg: 143.22 / Max: 157.87 Min: 17.52 / Avg: 128.33 / Max: 140.79 Min: 28.56 / Avg: 270.92 / Max: 332.67 Min: 59.6 / Avg: 495.69 / Max: 689.03 Min: 33.17 / Avg: 309.02 / Max: 409.54 Min: 68.17 / Avg: 526.7 / Max: 819.28 Min: 21.25 / Avg: 249.04 / Max: 321.21 Min: 42.16 / Avg: 450.48 / Max: 658.58 Min: 72.8 / Avg: 328.51 / Max: 377.04 Min: 197.55 / Avg: 624.98 / Max: 753.94 Min: 83.35 / Avg: 329.11 / Max: 376.3 Min: 213.51 / Avg: 619.95 / Max: 751.65 Min: 59.23 / Avg: 321.19 / Max: 377.05 Min: 168.54 / Avg: 600.33 / Max: 753.81 Min: 33.32 / Avg: 313.77 / Max: 375.72 Min: 82.45 / Avg: 561.7 / Max: 752.76
Primesieve Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 12.1 Length: 1e13 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 15 30 45 60 75 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 4 SE +/- 0.02, N = 3 SE +/- 0.02, N = 4 SE +/- 0.02, N = 3 SE +/- 0.01, N = 4 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 45.95 54.39 24.24 12.63 23.27 12.13 21.76 11.56 67.32 33.92 58.16 30.20 51.82 27.11 48.41 25.02 1. (CXX) g++ options: -O3 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Primesieve 12.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.27 / Avg: 151.7 / Max: 165.81 Min: 15.99 / Avg: 116.97 / Max: 130.2 Min: 28.63 / Avg: 313.43 / Max: 370.81 Min: 38.5 / Avg: 568.09 / Max: 750.08 Min: 34.15 / Avg: 347.93 / Max: 410.94 Min: 60.51 / Avg: 618.54 / Max: 821.49 Min: 21.63 / Avg: 274.16 / Max: 334.73 Min: 41.03 / Avg: 500.8 / Max: 682.66 Min: 74.26 / Avg: 333.11 / Max: 377.38 Min: 198.12 / Avg: 644.29 / Max: 754.21 Min: 74.48 / Avg: 332.68 / Max: 375.62 Min: 195.98 / Avg: 641.65 / Max: 754.11 Min: 54.67 / Avg: 328.05 / Max: 377.53 Min: 169.87 / Avg: 623.11 / Max: 754.26 Min: 21.47 / Avg: 325.15 / Max: 377.46 Min: 72.85 / Avg: 610.7 / Max: 755.46
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Person Detection FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 0.19, N = 3 SE +/- 0.41, N = 3 SE +/- 2.90, N = 15 SE +/- 0.44, N = 15 SE +/- 1.22, N = 15 SE +/- 0.09, N = 3 SE +/- 6.97, N = 15 SE +/- 1.90, N = 15 SE +/- 0.18, N = 3 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 161.48 156.59 157.45 63.98 110.91 43.77 230.24 85.86 41.93 42.16 59.69 64.91 100.39 45.78 39.62 43.41 MIN: 55.89 / MAX: 223.7 MIN: 61.92 / MAX: 224.56 MIN: 52.3 / MAX: 282.08 MIN: 39.78 / MAX: 114.18 MIN: 43.17 / MAX: 198.04 MIN: 37.7 / MAX: 90.56 MIN: 65.18 / MAX: 718.24 MIN: 47.19 / MAX: 207.44 MIN: 27.14 / MAX: 54.05 MIN: 32.47 / MAX: 70.84 MIN: 42.86 / MAX: 84.25 MIN: 49.88 / MAX: 107.03 MIN: 41.76 / MAX: 139.86 MIN: 33.94 / MAX: 85.21 MIN: 26.71 / MAX: 62.24 MIN: 27.3 / MAX: 82.85 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.75 / Avg: 155.49 / Max: 170.23 Min: 16.08 / Avg: 138.77 / Max: 151.09 Min: 28.22 / Avg: 304.28 / Max: 347.63 Min: 55.49 / Avg: 628.28 / Max: 705.07 Min: 34.13 / Avg: 363.03 / Max: 408.76 Min: 69.94 / Avg: 750.87 / Max: 819.03 Min: 20.51 / Avg: 267.71 / Max: 344.82 Min: 40.79 / Avg: 578.55 / Max: 688.18 Min: 75.73 / Avg: 330.51 / Max: 371.12 Min: 197.25 / Avg: 664.45 / Max: 750.81 Min: 74.3 / Avg: 331.92 / Max: 379.36 Min: 198.61 / Avg: 661.77 / Max: 760.61 Min: 59.06 / Avg: 327.33 / Max: 376.83 Min: 166.26 / Avg: 657.43 / Max: 746.56 Min: 34.28 / Avg: 219.81 / Max: 297.36 Min: 83.28 / Avg: 525.5 / Max: 710.63
Llamafile Mozilla's Llamafile allows distributing and running large language models (LLMs) as a single file. Llamafile aims to make open-source LLMs more accessible to developers and users. Llamafile supports a variety of models, CPUs and GPUs, and other options. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.7 Test: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.12, N = 6 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 5.16 5.11 8.35 5.24 8.48 5.24 7.47 4.63 1.47 1.77 1.54 1.54 5.85 3.82 4.08 3.05 Tokens Per Second Per Watt
OpenBenchmarking.org Tokens Per Second Per Watt, More Is Better Llamafile 0.7 Test: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0099 0.0198 0.0297 0.0396 0.0495 0.038 0.044 0.034 0.011 0.029 0.009 0.037 0.011 0.005 0.003 0.005 0.002 0.018 0.006 0.013 0.005 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Llamafile 0.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.14 / Avg: 135.55 / Max: 164.93 Min: 16.37 / Avg: 115.77 / Max: 139.52 Min: 30.07 / Avg: 245.28 / Max: 317.66 Min: 58.45 / Avg: 480.08 / Max: 600.8 Min: 29.28 / Avg: 296.68 / Max: 399.75 Min: 65.77 / Avg: 568.13 / Max: 708.34 Min: 21.63 / Avg: 203.84 / Max: 298.7 Min: 41.26 / Avg: 412.48 / Max: 564 Min: 72.56 / Avg: 306.79 / Max: 356.97 Min: 203.4 / Avg: 633.28 / Max: 704.68 Min: 86.22 / Avg: 325.29 / Max: 362.03 Min: 184.95 / Avg: 639.77 / Max: 696.32 Min: 53.92 / Avg: 317.1 / Max: 376.78 Min: 170.34 / Avg: 630.52 / Max: 723.74 Min: 33.89 / Avg: 307.46 / Max: 338.87 Min: 85.64 / Avg: 568.91 / Max: 665.68
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.3 Algorithm: ChaCha20 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300000M 600000M 900000M 1200000M 1500000M SE +/- 9570566.28, N = 3 SE +/- 71110585.03, N = 3 SE +/- 151272595.84, N = 3 SE +/- 80233613.02, N = 3 SE +/- 94935214.36, N = 3 SE +/- 874940734.18, N = 3 SE +/- 170968063.89, N = 3 SE +/- 63368290.15, N = 3 SE +/- 415487944.57, N = 3 SE +/- 329510485.23, N = 3 SE +/- 312824840.64, N = 3 SE +/- 863978337.15, N = 3 SE +/- 232288636.76, N = 3 SE +/- 271318554.29, N = 3 SE +/- 3335467.26, N = 3 SE +/- 303075210.03, N = 3 307068938377 282053327077 503311021123 1001502634860 544904084773 1086618191950 656396329387 1306928507573 226597354153 449908494927 263740690067 522912403347 316126421173 624402972947 344150916937 684284382760 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.3 Algorithm: ChaCha20 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 400M 800M 1200M 1600M 2000M 1824186596.16 2009755510.33 1510491450.86 1480134319.02 1361592598.33 1353849347.71 1795036060.88 1778324265.62 811931853.54 774438543.71 852735838.29 846331621.99 977212174.27 936302176.04 1008751851.93 1003586866.34 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 150 300 450 600 750 Min: 17.59 / Avg: 168.33 / Max: 176.1 Min: 15.27 / Avg: 140.34 / Max: 147.74 Min: 30.31 / Avg: 333.21 / Max: 358.89 Min: 59.94 / Avg: 676.63 / Max: 728.84 Min: 34.55 / Avg: 400.2 / Max: 411.05 Min: 70.23 / Avg: 802.61 / Max: 826.12 Min: 22.28 / Avg: 365.67 / Max: 391.73 Min: 42.15 / Avg: 734.92 / Max: 800.62 Min: 71.67 / Avg: 279.08 / Max: 326.73 Min: 197.94 / Avg: 580.95 / Max: 674.82 Min: 73.21 / Avg: 309.29 / Max: 373.85 Min: 183.97 / Avg: 617.86 / Max: 744.83 Min: 54.19 / Avg: 323.5 / Max: 377.22 Min: 168.89 / Avg: 666.88 / Max: 755.69 Min: 22.69 / Avg: 341.17 / Max: 376.63 Min: 70.13 / Avg: 681.84 / Max: 728.96
Xcompact3d Incompact3d Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 193 Cells Per Direction EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.01040414, N = 3 SE +/- 0.10237421, N = 3 SE +/- 0.06577912, N = 5 SE +/- 0.04312156, N = 5 SE +/- 0.05847189, N = 6 SE +/- 0.00891960, N = 5 SE +/- 0.03951909, N = 5 SE +/- 0.01633300, N = 4 SE +/- 0.04445674, N = 4 SE +/- 0.01741826, N = 5 SE +/- 0.01708632, N = 4 SE +/- 0.05695382, N = 15 SE +/- 0.01004100, N = 4 SE +/- 0.04312801, N = 5 SE +/- 0.02923582, N = 4 SE +/- 0.03329814, N = 5 20.93791390 21.17803070 8.25383511 4.26078596 6.88121279 3.75319901 8.87225914 4.51061893 13.96923760 7.29349117 12.79975820 6.66387510 12.92934130 6.39169817 10.19670660 5.39125967 1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.02 / Avg: 124.9 / Max: 149.76 Min: 17.28 / Avg: 114.52 / Max: 138.44 Min: 30.56 / Avg: 216.15 / Max: 325.7 Min: 57.92 / Avg: 371 / Max: 659.72 Min: 34.54 / Avg: 254.46 / Max: 403.7 Min: 64.22 / Avg: 431.35 / Max: 814.6 Min: 22.21 / Avg: 200.17 / Max: 312.99 Min: 40.05 / Avg: 313.08 / Max: 632.39 Min: 75.94 / Avg: 288.56 / Max: 366.2 Min: 187.55 / Avg: 524.04 / Max: 746.57 Min: 77.21 / Avg: 283.32 / Max: 372.73 Min: 196.28 / Avg: 524.59 / Max: 749.74 Min: 58.88 / Avg: 278.29 / Max: 373.12 Min: 151.7 / Avg: 495.02 / Max: 746.22 Min: 24.14 / Avg: 251.56 / Max: 369.25 Min: 78.61 / Avg: 426.75 / Max: 741.39
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.3 Algorithm: ChaCha20-Poly1305 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200000M 400000M 600000M 800000M 1000000M SE +/- 24234601.89, N = 3 SE +/- 65233629.47, N = 3 SE +/- 29295317.86, N = 3 SE +/- 6119305.11, N = 3 SE +/- 4478378.70, N = 3 SE +/- 558033954.71, N = 3 SE +/- 151303743.70, N = 3 SE +/- 19082587.00, N = 3 SE +/- 6370607.14, N = 3 SE +/- 4148679.28, N = 3 SE +/- 37795783.50, N = 3 SE +/- 36941390.18, N = 3 SE +/- 766713911.50, N = 3 SE +/- 132227134.30, N = 3 SE +/- 120899798.30, N = 3 SE +/- 82161991.11, N = 3 211185661727 193305093643 350960998357 698925822223 383925652077 767652719547 457039748870 906028608990 161321058213 320396720447 187796174440 372735215297 225038850730 442202532977 223625738013 444099144270 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.3 Algorithm: ChaCha20-Poly1305 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300M 600M 900M 1200M 1500M 1287300050.02 1388893901.10 1061980440.86 1039886433.31 957073830.82 957228032.13 1269363721.24 1244610923.33 568545794.25 542055990.77 603664493.39 574875604.57 666607882.23 644676877.17 658820863.84 648595124.99 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.74 / Avg: 164.05 / Max: 171.53 Min: 14.85 / Avg: 139.18 / Max: 148.57 Min: 27.82 / Avg: 330.48 / Max: 353.86 Min: 54.76 / Avg: 672.12 / Max: 721.51 Min: 35.8 / Avg: 401.15 / Max: 411.05 Min: 68.91 / Avg: 801.95 / Max: 822.13 Min: 21.97 / Avg: 360.05 / Max: 375.06 Min: 41.79 / Avg: 727.96 / Max: 758.79 Min: 74.91 / Avg: 283.74 / Max: 304.03 Min: 186.72 / Avg: 591.08 / Max: 630.87 Min: 71.53 / Avg: 311.09 / Max: 362.15 Min: 193.92 / Avg: 648.38 / Max: 695.92 Min: 58.01 / Avg: 337.59 / Max: 352.56 Min: 163.33 / Avg: 685.93 / Max: 736.16 Min: 21.83 / Avg: 339.43 / Max: 376.89 Min: 79.72 / Avg: 684.71 / Max: 754.83
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Weld Porosity Detection FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.08, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 4 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 11.78 13.11 10.09 10.17 9.42 9.25 10.92 11.20 2.50 2.53 2.67 2.79 3.58 2.50 2.36 2.48 MIN: 6.2 / MAX: 24.38 MIN: 6.01 / MAX: 25.62 MIN: 4.29 / MAX: 26.52 MIN: 8.74 / MAX: 41.95 MIN: 4.22 / MAX: 26.07 MIN: 8.35 / MAX: 44.99 MIN: 4.98 / MAX: 37.58 MIN: 9.95 / MAX: 64.26 MIN: 2.22 / MAX: 27.89 MIN: 2.23 / MAX: 31.44 MIN: 2.39 / MAX: 34.22 MIN: 2.44 / MAX: 34.21 MIN: 1.4 / MAX: 12.71 MIN: 2.28 / MAX: 40.8 MIN: 2.02 / MAX: 21.33 MIN: 1.98 / MAX: 34.01 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.69 / Avg: 156.99 / Max: 167.88 Min: 17.61 / Avg: 128.88 / Max: 137.57 Min: 28.34 / Avg: 326.4 / Max: 351.05 Min: 59.07 / Avg: 649.54 / Max: 702.57 Min: 34.9 / Avg: 384.48 / Max: 410.25 Min: 68.11 / Avg: 762.39 / Max: 819.68 Min: 21.95 / Avg: 308.04 / Max: 333.19 Min: 40.68 / Avg: 615.84 / Max: 667.35 Min: 73.64 / Avg: 331.47 / Max: 377.08 Min: 208.01 / Avg: 666.11 / Max: 753.96 Min: 83.9 / Avg: 332.66 / Max: 375.98 Min: 206.25 / Avg: 666.2 / Max: 754.51 Min: 58.8 / Avg: 329.61 / Max: 364.22 Min: 168.91 / Avg: 662.23 / Max: 753.52 Min: 34.78 / Avg: 327.29 / Max: 381.26 Min: 83.6 / Avg: 652.57 / Max: 760.35
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Person Detection FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200 400 600 800 1000 SE +/- 0.25, N = 3 SE +/- 0.56, N = 3 SE +/- 7.13, N = 15 SE +/- 5.47, N = 15 SE +/- 5.20, N = 15 SE +/- 2.17, N = 3 SE +/- 13.40, N = 15 SE +/- 20.94, N = 15 SE +/- 1.23, N = 3 SE +/- 1.28, N = 3 SE +/- 0.52, N = 3 SE +/- 0.55, N = 3 SE +/- 0.04, N = 3 SE +/- 0.69, N = 3 SE +/- 0.95, N = 3 SE +/- 1.42, N = 3 197.91 204.05 306.27 750.11 433.06 1095.79 283.18 750.96 285.95 568.81 301.30 569.60 298.56 654.84 403.20 736.23 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
Speedb Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better Speedb 2.7 Test: Random Read EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300M 600M 900M 1200M 1500M SE +/- 109015.29, N = 3 SE +/- 70537.88, N = 3 SE +/- 66268.38, N = 3 SE +/- 955915.20, N = 3 SE +/- 205170.43, N = 3 SE +/- 5954964.38, N = 3 SE +/- 1606748.01, N = 3 SE +/- 2399164.12, N = 3 SE +/- 3698537.65, N = 15 SE +/- 8491098.40, N = 15 SE +/- 2763560.64, N = 13 SE +/- 2955608.21, N = 3 SE +/- 168680.90, N = 3 SE +/- 5393159.45, N = 3 SE +/- 242146.75, N = 3 SE +/- 5399265.88, N = 15 274091778 242940110 428557054 857687179 537633034 1059546533 589951070 1169947385 211457048 419943570 244070674 497907490 276179617 551286958 315178381 586147871 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Speedb 2.7 Test: Random Read EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 400K 800K 1200K 1600K 2000K 1818446.43 2023204.53 1563545.14 1511464.50 1404496.49 1380809.63 1991742.68 1962564.72 680456.82 644065.87 744809.12 747196.35 834970.47 834394.35 962315.70 894980.08 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Speedb 2.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.42 / Avg: 150.73 / Max: 161.26 Min: 18.13 / Avg: 120.08 / Max: 129.07 Min: 29.2 / Avg: 274.09 / Max: 293.61 Min: 60.4 / Avg: 567.45 / Max: 609.28 Min: 33.89 / Avg: 382.79 / Max: 411.07 Min: 69.56 / Avg: 767.34 / Max: 822.12 Min: 22.11 / Avg: 296.2 / Max: 318.84 Min: 40.77 / Avg: 596.13 / Max: 648.34 Min: 71.79 / Avg: 310.76 / Max: 333.42 Min: 185.93 / Avg: 652.02 / Max: 753.16 Min: 70.78 / Avg: 327.7 / Max: 377.13 Min: 196.64 / Avg: 666.37 / Max: 752.39 Min: 54.25 / Avg: 330.77 / Max: 378.66 Min: 168.96 / Avg: 660.7 / Max: 755.23 Min: 27.16 / Avg: 327.52 / Max: 377.21 Min: 68.16 / Avg: 654.93 / Max: 718.87
RocksDB This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better RocksDB 9.0 Test: Random Read EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200M 400M 600M 800M 1000M SE +/- 206014.27, N = 3 SE +/- 211259.57, N = 3 SE +/- 610621.13, N = 3 SE +/- 599126.55, N = 3 SE +/- 326038.01, N = 3 SE +/- 2927744.32, N = 3 SE +/- 153963.59, N = 3 SE +/- 1154820.73, N = 3 SE +/- 81327.78, N = 3 SE +/- 1120857.00, N = 3 SE +/- 86315.43, N = 3 SE +/- 785893.13, N = 3 SE +/- 3338569.32, N = 3 SE +/- 1411218.60, N = 3 SE +/- 1119855.47, N = 3 SE +/- 82996.27, N = 3 266186537 234992636 409272949 815145110 523310091 1031209151 573235452 1135206564 205346154 405533173 234469939 473755946 265366170 525049998 297517946 578392800 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better RocksDB 9.0 Test: Random Read EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 400K 800K 1200K 1600K 2000K 1780768.80 1972313.57 1523041.95 1477897.34 1362651.04 1341710.00 1957165.94 1907427.82 663946.02 608125.97 704496.50 709419.37 802721.25 791368.94 904750.01 878232.54 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better RocksDB 9.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.07 / Avg: 149.48 / Max: 159.86 Min: 18.15 / Avg: 119.15 / Max: 127.37 Min: 24.72 / Avg: 268.72 / Max: 291.05 Min: 53.06 / Avg: 551.56 / Max: 593.87 Min: 33.04 / Avg: 384.04 / Max: 411.06 Min: 61.31 / Avg: 768.58 / Max: 822.92 Min: 21.45 / Avg: 292.89 / Max: 315.65 Min: 40.42 / Avg: 595.15 / Max: 641.39 Min: 73.12 / Avg: 309.28 / Max: 327.48 Min: 186.94 / Avg: 666.86 / Max: 752.6 Min: 77.85 / Avg: 332.82 / Max: 365.47 Min: 195.53 / Avg: 667.81 / Max: 755.48 Min: 54.21 / Avg: 330.58 / Max: 378.42 Min: 145.71 / Avg: 663.47 / Max: 754.3 Min: 25.38 / Avg: 328.84 / Max: 377.98 Min: 63.13 / Avg: 658.59 / Max: 754.12
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Person Re-Identification Retail FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4K 8K 12K 16K 20K SE +/- 3.18, N = 3 SE +/- 1.89, N = 3 SE +/- 4.14, N = 3 SE +/- 7.03, N = 3 SE +/- 4.58, N = 3 SE +/- 7.99, N = 3 SE +/- 4.98, N = 3 SE +/- 10.13, N = 3 SE +/- 3.72, N = 3 SE +/- 3.67, N = 3 SE +/- 0.46, N = 3 SE +/- 7.22, N = 3 SE +/- 3.97, N = 3 SE +/- 4.14, N = 3 SE +/- 5.57, N = 3 SE +/- 7.74, N = 3 3692.67 3355.73 6554.12 12024.18 6846.79 12665.62 7862.65 14601.52 6717.65 13102.28 7207.74 14033.94 5943.66 15519.92 9515.86 18375.63 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Road Segmentation ADAS FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 15 30 45 60 75 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.26, N = 15 SE +/- 0.02, N = 3 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 SE +/- 0.21, N = 3 SE +/- 0.06, N = 3 SE +/- 0.14, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.28, N = 3 31.99 34.76 27.46 14.30 21.63 11.95 38.21 15.84 55.74 56.71 62.53 62.11 55.99 65.02 55.07 57.13 MIN: 16.09 / MAX: 51.04 MIN: 17.72 / MAX: 54.79 MIN: 14.03 / MAX: 53.11 MIN: 11.91 / MAX: 38.94 MIN: 14.35 / MAX: 50.29 MIN: 11.25 / MAX: 37.16 MIN: 15.81 / MAX: 88.01 MIN: 13.68 / MAX: 55.05 MIN: 44.83 / MAX: 69.99 MIN: 45.43 / MAX: 89.46 MIN: 48.23 / MAX: 78.43 MIN: 47.82 / MAX: 101.26 MIN: 27.63 / MAX: 76.45 MIN: 55.15 / MAX: 104.43 MIN: 46.78 / MAX: 80.73 MIN: 39.58 / MAX: 104.91 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.37 / Avg: 152.34 / Max: 164.44 Min: 17.95 / Avg: 130.31 / Max: 141.74 Min: 30.63 / Avg: 302.4 / Max: 328.21 Min: 60.61 / Avg: 599.17 / Max: 650.25 Min: 33.63 / Avg: 370.93 / Max: 400.44 Min: 67.8 / Avg: 760.09 / Max: 819.06 Min: 21.36 / Avg: 269.44 / Max: 322.88 Min: 40.64 / Avg: 591.23 / Max: 648.1 Min: 74.41 / Avg: 330.18 / Max: 376.61 Min: 200.42 / Avg: 659.67 / Max: 754.38 Min: 84.64 / Avg: 330.18 / Max: 380.81 Min: 201 / Avg: 659.95 / Max: 753.12 Min: 59.15 / Avg: 327.9 / Max: 376.83 Min: 168.6 / Avg: 652.87 / Max: 754.8 Min: 34.63 / Avg: 325.47 / Max: 376.24 Min: 81.23 / Avg: 641.35 / Max: 755.31
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: WPA PSK EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300K 600K 900K 1200K 1500K SE +/- 321.00, N = 3 SE +/- 79.20, N = 3 SE +/- 921.74, N = 3 SE +/- 1333.33, N = 3 SE +/- 2683.36, N = 3 SE +/- 5291.50, N = 3 SE +/- 478.70, N = 3 SE +/- 4358.90, N = 3 SE +/- 987.33, N = 3 SE +/- 5122.51, N = 3 SE +/- 1274.00, N = 3 SE +/- 788.50, N = 3 SE +/- 1599.31, N = 3 SE +/- 2519.00, N = 3 SE +/- 864.17, N = 3 SE +/- 15579.94, N = 15 386212 339998 643475 1253333 671322 1294000 803187 1551000 286552 562875 334076 657889 401278 793253 428116 797093 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2 Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 2023.03.14 Test: WPA PSK EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 600 1200 1800 2400 3000 2414.27 2563.93 2017.72 1941.48 1853.75 1799.50 2304.83 2253.10 1133.97 1057.22 1204.61 1100.83 1347.41 1261.48 1374.72 1285.88 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better John The Ripper 2023.03.14 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.24 / Avg: 159.97 / Max: 182.75 Min: 17.08 / Avg: 132.61 / Max: 150.07 Min: 29.6 / Avg: 318.91 / Max: 365.41 Min: 59 / Avg: 645.56 / Max: 737.44 Min: 33.28 / Avg: 362.14 / Max: 411.02 Min: 67.44 / Avg: 719.09 / Max: 821.15 Min: 22.59 / Avg: 348.48 / Max: 397.44 Min: 43.51 / Avg: 688.39 / Max: 796.51 Min: 66.96 / Avg: 252.7 / Max: 281.63 Min: 186.18 / Avg: 532.41 / Max: 591.91 Min: 76.35 / Avg: 277.33 / Max: 310.37 Min: 193.96 / Avg: 597.63 / Max: 753.44 Min: 56.65 / Avg: 297.81 / Max: 364.7 Min: 165.05 / Avg: 628.83 / Max: 743.61 Min: 23.24 / Avg: 311.42 / Max: 376.46 Min: 64.41 / Avg: 619.88 / Max: 753.08
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.3 Algorithm: RSA4096 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 800K 1600K 2400K 3200K 4000K SE +/- 126.84, N = 3 SE +/- 136.07, N = 3 SE +/- 149.48, N = 3 SE +/- 5.26, N = 3 SE +/- 1576.96, N = 3 SE +/- 6125.37, N = 3 SE +/- 444.74, N = 3 SE +/- 2222.55, N = 3 SE +/- 165.80, N = 3 SE +/- 355.69, N = 3 SE +/- 200.66, N = 3 SE +/- 277.84, N = 3 SE +/- 92.46, N = 3 SE +/- 862.96, N = 3 SE +/- 92.62, N = 3 SE +/- 6745.42, N = 3 881480.8 810291.8 1436324.9 2861959.8 1613828.7 3214376.3 1872096.1 3729470.8 697707.2 1395446.4 813667.6 1627941.4 974189.9 1916162.9 1036735.6 2057021.6 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl verify/s Per Watt
OpenBenchmarking.org verify/s Per Watt, More Is Better OpenSSL 3.3 Algorithm: RSA4096 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1400 2800 4200 5600 7000 5691.68 6595.71 4919.81 4728.28 4121.23 4099.33 5836.73 5696.67 2306.66 2215.41 2489.80 2430.79 2907.57 2791.77 3030.66 2997.55 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 15.75 / Avg: 154.87 / Max: 162.9 Min: 16.2 / Avg: 122.85 / Max: 135.57 Min: 27.6 / Avg: 291.95 / Max: 323.61 Min: 59.69 / Avg: 605.29 / Max: 659.72 Min: 32.96 / Avg: 391.59 / Max: 411.04 Min: 64.61 / Avg: 784.12 / Max: 822.1 Min: 19.93 / Avg: 320.74 / Max: 351.44 Min: 43.66 / Avg: 654.68 / Max: 716.5 Min: 73.89 / Avg: 302.47 / Max: 335.7 Min: 198.5 / Avg: 629.88 / Max: 694.83 Min: 74.71 / Avg: 326.8 / Max: 367.81 Min: 199.02 / Avg: 669.72 / Max: 754.93 Min: 56.91 / Avg: 335.05 / Max: 351.44 Min: 165.7 / Avg: 686.36 / Max: 737.53 Min: 32.63 / Avg: 342.08 / Max: 377.97 Min: 70.64 / Avg: 686.24 / Max: 753.6
RELION RELION - REgularised LIkelihood OptimisatioN - is a stand-alone computer program for Maximum A Posteriori refinement of (multiple) 3D reconstructions or 2D class averages in cryo-electron microscopy (cryo-EM). It is developed in the research group of Sjors Scheres at the MRC Laboratory of Molecular Biology. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better RELION 4.0.1 Test: Basic - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 SE +/- 4.63, N = 3 SE +/- 6.20, N = 3 SE +/- 2.95, N = 3 SE +/- 2.11, N = 12 SE +/- 2.17, N = 3 SE +/- 1.27, N = 15 SE +/- 7.69, N = 9 SE +/- 4.04, N = 12 SE +/- 2.98, N = 3 SE +/- 3.24, N = 12 SE +/- 4.65, N = 9 SE +/- 2.66, N = 12 SE +/- 7.14, N = 9 SE +/- 2.65, N = 3 SE +/- 2.68, N = 7 SE +/- 1.41, N = 3 468.69 496.24 238.88 145.65 176.42 93.17 360.21 201.27 412.37 214.42 384.38 201.38 373.18 203.41 290.45 155.74 1. (CXX) g++ options: -fopenmp -std=c++11 -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -ljpeg -lmpi_cxx -lmpi CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better RELION 4.0.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.83 / Avg: 163.61 / Max: 166.72 Min: 17.41 / Avg: 148.71 / Max: 151.06 Min: 28.15 / Avg: 320.59 / Max: 347.29 Min: 52.45 / Avg: 622.74 / Max: 703.72 Min: 32.42 / Avg: 394.87 / Max: 409.71 Min: 63.79 / Avg: 752.52 / Max: 819.67 Min: 21.86 / Avg: 270.75 / Max: 336.67 Min: 42.44 / Avg: 553.69 / Max: 696.76 Min: 73.55 / Avg: 345.11 / Max: 376.88 Min: 199.08 / Avg: 683.82 / Max: 754.9 Min: 73.36 / Avg: 344.98 / Max: 377.25 Min: 195.56 / Avg: 682.36 / Max: 754.93 Min: 50.74 / Avg: 344.24 / Max: 377.4 Min: 170.14 / Avg: 679.44 / Max: 754.54 Min: 22.57 / Avg: 341.82 / Max: 377.61 Min: 77.4 / Avg: 665.17 / Max: 755.23
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Vehicle Detection FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.12, N = 15 SE +/- 0.00, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 10.78 11.48 8.58 4.46 7.85 4.08 10.97 5.00 16.33 17.19 19.44 20.24 17.84 21.73 13.73 14.67 MIN: 5.12 / MAX: 22.93 MIN: 5.61 / MAX: 26.2 MIN: 3.84 / MAX: 27.82 MIN: 3.61 / MAX: 26.57 MIN: 3.77 / MAX: 30.51 MIN: 3.61 / MAX: 25.59 MIN: 4.5 / MAX: 46.56 MIN: 4.38 / MAX: 33.93 MIN: 13.49 / MAX: 26.98 MIN: 13.16 / MAX: 42.02 MIN: 15.1 / MAX: 32.19 MIN: 13.69 / MAX: 36.9 MIN: 9.08 / MAX: 37.11 MIN: 15.08 / MAX: 39.82 MIN: 12.23 / MAX: 23.01 MIN: 12.73 / MAX: 41.28 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.62 / Avg: 156.89 / Max: 169.02 Min: 17.95 / Avg: 132.64 / Max: 143.89 Min: 29.58 / Avg: 315.71 / Max: 340.19 Min: 60.42 / Avg: 615.26 / Max: 666.72 Min: 34.99 / Avg: 371.14 / Max: 399.65 Min: 68.28 / Avg: 749.07 / Max: 815.77 Min: 21.25 / Avg: 305.35 / Max: 334.91 Min: 43.09 / Avg: 614.49 / Max: 665.93 Min: 71.96 / Avg: 331 / Max: 377.8 Min: 209.95 / Avg: 663.67 / Max: 754.1 Min: 75.87 / Avg: 332.1 / Max: 380.66 Min: 203.39 / Avg: 666.38 / Max: 758.84 Min: 59.46 / Avg: 330.49 / Max: 383.24 Min: 169.19 / Avg: 657.97 / Max: 750.83 Min: 33.71 / Avg: 302.37 / Max: 379.73 Min: 85.37 / Avg: 650.24 / Max: 758.93
QuantLib QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Multi-Threaded EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 160K 320K 480K 640K 800K SE +/- 82.80, N = 3 SE +/- 86.28, N = 3 SE +/- 86.78, N = 3 SE +/- 713.18, N = 3 SE +/- 412.72, N = 3 SE +/- 1065.76, N = 3 SE +/- 156.35, N = 3 SE +/- 804.57, N = 3 SE +/- 155.19, N = 3 SE +/- 286.86, N = 3 SE +/- 40.48, N = 3 SE +/- 160.59, N = 3 SE +/- 105.60, N = 3 SE +/- 362.38, N = 3 SE +/- 63.14, N = 3 SE +/- 248.00, N = 3 172566.8 153920.8 294183.2 593246.9 319827.2 644091.0 374413.5 759922.1 143755.3 288129.1 162727.4 316359.2 179647.5 350187.5 195648.8 380867.5 1. (CXX) g++ options: -O3 -march=native -fPIE -pie MFLOPS Per Watt
OpenBenchmarking.org MFLOPS Per Watt, More Is Better QuantLib 1.32 Configuration: Multi-Threaded EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300 600 900 1200 1500 1092.36 1202.80 930.70 921.01 864.58 881.58 1200.71 1200.40 447.64 436.19 495.61 477.83 552.74 537.18 605.78 585.75 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better QuantLib 1.32 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.59 / Avg: 157.98 / Max: 175.08 Min: 16.19 / Avg: 127.97 / Max: 140.32 Min: 28.93 / Avg: 316.09 / Max: 352.6 Min: 55.66 / Avg: 644.13 / Max: 715.22 Min: 33.64 / Avg: 369.92 / Max: 411.04 Min: 65.34 / Avg: 730.61 / Max: 822.15 Min: 21.13 / Avg: 311.83 / Max: 350.73 Min: 40.73 / Avg: 633.06 / Max: 715.84 Min: 72.72 / Avg: 321.14 / Max: 375.61 Min: 205.33 / Avg: 660.57 / Max: 754.82 Min: 83.08 / Avg: 328.34 / Max: 376.88 Min: 203.63 / Avg: 662.07 / Max: 755.13 Min: 58.59 / Avg: 325.01 / Max: 377.73 Min: 167.37 / Avg: 651.9 / Max: 756.65 Min: 31.57 / Avg: 322.97 / Max: 378.63 Min: 77.95 / Avg: 650.23 / Max: 756.48
m-queens A solver for the N-queens problem with multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.019, N = 4 SE +/- 0.029, N = 4 SE +/- 0.019, N = 6 SE +/- 0.093, N = 15 SE +/- 0.025, N = 6 SE +/- 0.064, N = 15 SE +/- 0.021, N = 6 SE +/- 0.031, N = 8 SE +/- 0.031, N = 3 SE +/- 0.021, N = 5 SE +/- 0.035, N = 3 SE +/- 0.028, N = 5 SE +/- 0.025, N = 4 SE +/- 0.021, N = 6 SE +/- 0.030, N = 4 SE +/- 0.028, N = 6 13.250 14.493 7.738 4.609 6.950 4.133 6.350 3.981 21.030 10.840 18.102 9.285 15.211 7.881 14.257 7.402 1. (CXX) g++ options: -fopenmp -O2 -march=native CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better m-queens 1.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.39 / Avg: 122.5 / Max: 158.25 Min: 15.51 / Avg: 106.99 / Max: 136.77 Min: 28.35 / Avg: 221.83 / Max: 330.6 Min: 52.18 / Avg: 366.51 / Max: 668.54 Min: 31.01 / Avg: 267.9 / Max: 411.07 Min: 63.38 / Avg: 412.8 / Max: 820.43 Min: 20.79 / Avg: 202.95 / Max: 326.11 Min: 41.4 / Avg: 316.2 / Max: 657.62 Min: 75.51 / Avg: 241.11 / Max: 276.68 Min: 201.32 / Avg: 465.55 / Max: 574.13 Min: 77.06 / Avg: 260.39 / Max: 302.34 Min: 180.86 / Avg: 502.47 / Max: 745.73 Min: 62.12 / Avg: 266.4 / Max: 324.62 Min: 165.88 / Avg: 497.84 / Max: 742.37 Min: 32.66 / Avg: 278.57 / Max: 374.34 Min: 77.88 / Avg: 481.61 / Max: 753.74
Coremark This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2M 4M 6M 8M 10M SE +/- 1374.66, N = 3 SE +/- 11678.22, N = 3 SE +/- 10132.66, N = 3 SE +/- 2306.54, N = 3 SE +/- 54326.71, N = 3 SE +/- 76743.24, N = 3 SE +/- 4775.53, N = 3 SE +/- 54771.53, N = 3 SE +/- 1747.99, N = 3 SE +/- 7219.73, N = 3 SE +/- 5202.86, N = 3 SE +/- 20649.43, N = 3 SE +/- 9256.47, N = 3 SE +/- 12743.41, N = 3 SE +/- 4635.95, N = 3 SE +/- 9700.21, N = 3 2274636.13 2069432.53 3926265.28 7975907.45 4152864.88 8448083.73 4525288.00 8171241.71 1614763.70 2993136.60 1855135.33 3279262.53 2194630.60 3902076.24 2369998.36 4966250.78 1. (CC) gcc options: -O2 -lrt" -lrt Iterations/Sec Per Watt
OpenBenchmarking.org Iterations/Sec Per Watt, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4K 8K 12K 16K 20K 15712.09 16836.69 13254.14 14883.36 12144.31 13681.18 16764.47 16699.88 6163.40 5763.20 6571.62 5880.62 7477.77 7118.47 7555.56 8414.46 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Coremark 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.44 / Avg: 144.77 / Max: 164.64 Min: 14.91 / Avg: 122.91 / Max: 138.78 Min: 26.88 / Avg: 296.23 / Max: 351.89 Min: 57.24 / Avg: 535.89 / Max: 711.09 Min: 32.2 / Avg: 341.96 / Max: 411.86 Min: 68.03 / Avg: 617.5 / Max: 820.53 Min: 21.14 / Avg: 269.93 / Max: 353.42 Min: 42.53 / Avg: 489.3 / Max: 675.15 Min: 76.41 / Avg: 261.99 / Max: 288.93 Min: 188.96 / Avg: 519.35 / Max: 613.18 Min: 77.43 / Avg: 282.3 / Max: 314.07 Min: 206.68 / Avg: 557.64 / Max: 672.47 Min: 54.25 / Avg: 293.49 / Max: 341.26 Min: 163.01 / Avg: 548.16 / Max: 709.21 Min: 34.34 / Avg: 313.68 / Max: 378.24 Min: 73.47 / Avg: 590.2 / Max: 753.92
Blender Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.1 Blend File: Classroom - Compute: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20 40 60 80 100 SE +/- 0.04, N = 3 SE +/- 0.18, N = 3 SE +/- 0.11, N = 3 SE +/- 0.05, N = 3 SE +/- 0.14, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.20, N = 3 SE +/- 0.38, N = 3 SE +/- 0.40, N = 3 SE +/- 0.25, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.22, N = 3 67.67 75.26 40.26 21.12 35.66 18.67 31.65 16.82 86.67 44.71 76.87 40.38 68.61 35.91 63.61 33.61 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Blender 4.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.44 / Avg: 159.04 / Max: 169.83 Min: 17.21 / Avg: 130.93 / Max: 139.49 Min: 31.02 / Avg: 300.94 / Max: 333.76 Min: 59.51 / Avg: 570.92 / Max: 684.42 Min: 32.63 / Avg: 365.01 / Max: 410.63 Min: 62.19 / Avg: 671.43 / Max: 821.27 Min: 21.92 / Avg: 295.34 / Max: 341.6 Min: 41.08 / Avg: 534.08 / Max: 683.35 Min: 77.25 / Avg: 334.43 / Max: 376.17 Min: 203.45 / Avg: 652.82 / Max: 753.03 Min: 77.77 / Avg: 335.02 / Max: 376.96 Min: 210.08 / Avg: 648.81 / Max: 752.1 Min: 60.22 / Avg: 330.03 / Max: 377.07 Min: 168.91 / Avg: 636.73 / Max: 754.11 Min: 24.56 / Avg: 326.8 / Max: 368.75 Min: 80.44 / Avg: 621.31 / Max: 724.97
Primesieve Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 12.1 Length: 1e12 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 SE +/- 0.006, N = 8 SE +/- 0.003, N = 8 SE +/- 0.003, N = 11 SE +/- 0.005, N = 12 SE +/- 0.004, N = 11 SE +/- 0.003, N = 12 SE +/- 0.002, N = 11 SE +/- 0.010, N = 15 SE +/- 0.006, N = 7 SE +/- 0.004, N = 9 SE +/- 0.010, N = 7 SE +/- 0.005, N = 10 SE +/- 0.019, N = 8 SE +/- 0.006, N = 10 SE +/- 0.007, N = 8 SE +/- 0.008, N = 11 3.801 4.545 2.058 1.242 1.984 1.199 1.921 1.359 6.160 3.121 5.332 2.691 4.576 2.325 4.250 2.187 1. (CXX) g++ options: -O3 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Primesieve 12.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.53 / Avg: 87.74 / Max: 161.22 Min: 16.77 / Avg: 72.3 / Max: 123.74 Min: 26.05 / Avg: 142.51 / Max: 369.88 Min: 58.4 / Avg: 217.76 / Max: 736.92 Min: 33.95 / Avg: 157.31 / Max: 410.23 Min: 65.08 / Avg: 239.55 / Max: 813.29 Min: 21.44 / Avg: 116.27 / Max: 317.57 Min: 39.28 / Avg: 188.39 / Max: 630.34 Min: 41.72 / Avg: 247.24 / Max: 378.01 Min: 185.19 / Avg: 432.51 / Max: 754.28 Min: 75.05 / Avg: 240.32 / Max: 376.42 Min: 188 / Avg: 419.03 / Max: 755.09 Min: 53.72 / Avg: 216.08 / Max: 377.44 Min: 167.18 / Avg: 384.89 / Max: 756.1 Min: 20.57 / Avg: 195.27 / Max: 377.63 Min: 70.6 / Avg: 304.92 / Max: 755.57
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.3 Algorithm: RSA4096 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20K 40K 60K 80K 100K SE +/- 22.17, N = 3 SE +/- 12.56, N = 3 SE +/- 29.11, N = 3 SE +/- 81.38, N = 3 SE +/- 98.71, N = 3 SE +/- 247.60, N = 3 SE +/- 27.42, N = 3 SE +/- 80.37, N = 3 SE +/- 19.45, N = 3 SE +/- 6.92, N = 3 SE +/- 20.13, N = 3 SE +/- 83.09, N = 3 SE +/- 9.84, N = 3 SE +/- 14.34, N = 3 SE +/- 15.15, N = 3 SE +/- 56.68, N = 3 25894.6 21047.5 35964.9 71942.0 48996.8 98237.0 53966.9 107830.6 22861.0 45531.0 25441.5 48534.8 27431.1 54601.1 31594.4 61499.6 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
OSPRay Studio Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 1.0 Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 9K 18K 27K 36K 45K SE +/- 250.73, N = 3 SE +/- 31.02, N = 3 SE +/- 44.33, N = 3 SE +/- 32.74, N = 3 SE +/- 39.30, N = 3 SE +/- 17.16, N = 3 SE +/- 33.33, N = 3 SE +/- 12.34, N = 3 SE +/- 28.06, N = 3 SE +/- 72.42, N = 3 SE +/- 83.47, N = 3 SE +/- 51.39, N = 3 SE +/- 62.46, N = 3 SE +/- 34.10, N = 3 SE +/- 37.49, N = 3 SE +/- 7.13, N = 3 38811 42513 18896 10428 16603 8916 15286 8413 37055 16645 33790 15357 31380 14138 23125 12124 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay Studio 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.86 / Avg: 118.24 / Max: 166.58 Min: 17.28 / Avg: 103.3 / Max: 141.73 Min: 29.03 / Avg: 265.51 / Max: 331.71 Min: 61.04 / Avg: 548.32 / Max: 683.37 Min: 35.54 / Avg: 320.39 / Max: 411.05 Min: 66.57 / Avg: 696.59 / Max: 822.15 Min: 21.92 / Avg: 245.02 / Max: 336.36 Min: 41.91 / Avg: 564.91 / Max: 692.23 Min: 74.46 / Avg: 283.63 / Max: 375.94 Min: 210.55 / Avg: 606.6 / Max: 753.89 Min: 78.1 / Avg: 281.43 / Max: 376.32 Min: 201.01 / Avg: 599.93 / Max: 754.22 Min: 61.9 / Avg: 259.12 / Max: 377.23 Min: 175.84 / Avg: 622.79 / Max: 756.66 Min: 35.35 / Avg: 299.63 / Max: 376.51 Min: 78.22 / Avg: 573.74 / Max: 753.21
Blender Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.1 Blend File: Pabellon Barcelona - Compute: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20 40 60 80 100 SE +/- 0.16, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 SE +/- 0.16, N = 3 SE +/- 0.11, N = 3 SE +/- 0.13, N = 3 SE +/- 0.13, N = 3 SE +/- 0.31, N = 3 SE +/- 0.08, N = 3 SE +/- 0.44, N = 3 SE +/- 0.20, N = 3 SE +/- 0.15, N = 3 SE +/- 0.26, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 85.12 94.41 48.88 26.35 44.87 23.84 39.83 21.66 106.96 55.26 97.80 51.29 87.92 46.38 80.29 41.85 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Blender 4.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.07 / Avg: 163.93 / Max: 175 Min: 17.1 / Avg: 133.17 / Max: 140.81 Min: 30.54 / Avg: 312.47 / Max: 346.36 Min: 58.1 / Avg: 592.5 / Max: 710.39 Min: 31.38 / Avg: 370.95 / Max: 410.68 Min: 65.23 / Avg: 677.05 / Max: 822.1 Min: 21.64 / Avg: 304.63 / Max: 348.18 Min: 41.28 / Avg: 465.54 / Max: 700.45 Min: 76.03 / Avg: 336.64 / Max: 376.9 Min: 211.91 / Avg: 656.12 / Max: 755.31 Min: 76.38 / Avg: 337.24 / Max: 375.65 Min: 208.34 / Avg: 654.26 / Max: 753.55 Min: 60.61 / Avg: 332.25 / Max: 377.75 Min: 172.77 / Avg: 638.84 / Max: 757.46 Min: 34.2 / Avg: 330.4 / Max: 377.77 Min: 77.37 / Avg: 624.76 / Max: 738.53
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.1 Blend File: Barbershop - Compute: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.25, N = 3 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.12, N = 3 SE +/- 0.23, N = 3 SE +/- 1.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.18, N = 3 SE +/- 0.42, N = 3 SE +/- 0.69, N = 3 SE +/- 0.21, N = 3 240.13 263.76 139.25 79.05 123.90 67.52 114.76 65.58 320.11 169.44 290.56 156.70 261.97 141.54 236.73 125.87 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Blender 4.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.05 / Avg: 166.92 / Max: 172.9 Min: 17.54 / Avg: 139.56 / Max: 146.03 Min: 27.59 / Avg: 319.36 / Max: 343.48 Min: 58.63 / Avg: 632.52 / Max: 694.54 Min: 35.25 / Avg: 388.09 / Max: 411.05 Min: 68.76 / Avg: 738.74 / Max: 823.72 Min: 23.25 / Avg: 321.43 / Max: 348.37 Min: 41.32 / Avg: 614.31 / Max: 703.89 Min: 73.43 / Avg: 343.06 / Max: 375.17 Min: 206.17 / Avg: 675.87 / Max: 751.49 Min: 85.64 / Avg: 342.89 / Max: 377.27 Min: 209.81 / Avg: 673.68 / Max: 752.4 Min: 60.47 / Avg: 340.87 / Max: 376.73 Min: 167.6 / Avg: 666.99 / Max: 755.27 Min: 33.95 / Avg: 339.43 / Max: 377.51 Min: 79.57 / Avg: 660.14 / Max: 755.17
OSPRay Studio Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 1.0 Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 10K 20K 30K 40K 50K SE +/- 111.72, N = 3 SE +/- 197.99, N = 3 SE +/- 5.55, N = 3 SE +/- 27.79, N = 3 SE +/- 24.66, N = 3 SE +/- 32.79, N = 3 SE +/- 21.05, N = 3 SE +/- 16.90, N = 3 SE +/- 36.47, N = 3 SE +/- 25.12, N = 3 SE +/- 74.33, N = 3 SE +/- 10.97, N = 3 SE +/- 61.20, N = 3 SE +/- 29.96, N = 3 SE +/- 26.00, N = 3 SE +/- 8.00, N = 3 45377 48844 22305 12427 19535 10514 18129 10015 42837 19701 39238 18111 36649 16837 31993 14358 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay Studio 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.33 / Avg: 121.97 / Max: 166.66 Min: 17.31 / Avg: 107.81 / Max: 142.82 Min: 27.99 / Avg: 273.01 / Max: 332.14 Min: 63.21 / Avg: 573.82 / Max: 678.02 Min: 35.75 / Avg: 329.8 / Max: 411.04 Min: 63.78 / Avg: 623.33 / Max: 822.16 Min: 22.55 / Avg: 256.1 / Max: 337.04 Min: 42.24 / Avg: 558.27 / Max: 695.79 Min: 68.84 / Avg: 291.26 / Max: 378.07 Min: 210.27 / Avg: 614.16 / Max: 754.39 Min: 76.02 / Avg: 287.68 / Max: 377.28 Min: 199.93 / Avg: 609.51 / Max: 752.77 Min: 59.63 / Avg: 268.15 / Max: 377.52 Min: 175.58 / Avg: 588.27 / Max: 757.42 Min: 26.59 / Avg: 252.77 / Max: 377.16 Min: 78.39 / Avg: 616.75 / Max: 753.89
PyTorch This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 14 28 42 56 70 SE +/- 0.18, N = 3 SE +/- 0.13, N = 3 SE +/- 0.21, N = 3 SE +/- 0.37, N = 12 SE +/- 0.09, N = 3 SE +/- 0.32, N = 12 SE +/- 0.08, N = 3 SE +/- 0.28, N = 12 SE +/- 0.76, N = 3 SE +/- 0.52, N = 3 SE +/- 0.62, N = 3 SE +/- 0.45, N = 3 SE +/- 0.52, N = 3 SE +/- 0.49, N = 3 SE +/- 0.64, N = 3 SE +/- 0.29, N = 3 33.68 33.30 34.90 17.41 34.90 16.97 27.78 13.11 52.97 41.68 51.56 37.91 54.84 37.44 63.60 41.45 MIN: 31.27 / MAX: 34.43 MIN: 32.26 / MAX: 33.94 MIN: 33.85 / MAX: 35.99 MIN: 11.49 / MAX: 19.87 MIN: 34.15 / MAX: 35.71 MIN: 14.73 / MAX: 19.36 MIN: 26.81 / MAX: 28.37 MIN: 8.01 / MAX: 14.98 MIN: 33.28 / MAX: 55.78 MIN: 37.38 / MAX: 44.02 MIN: 30.49 / MAX: 54.7 MIN: 16.15 / MAX: 40.53 MIN: 46.5 / MAX: 58.5 MIN: 26.88 / MAX: 41.26 MIN: 37.79 / MAX: 65.67 MIN: 38.42 / MAX: 42.93 batches/sec Per Watt
OpenBenchmarking.org batches/sec Per Watt, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0752 0.1504 0.2256 0.3008 0.376 0.305 0.334 0.179 0.048 0.150 0.040 0.176 0.044 0.197 0.073 0.189 0.066 0.218 0.077 0.293 0.082 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better PyTorch 2.2.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 Min: 19.91 / Avg: 110.56 / Max: 125.45 Min: 17.28 / Avg: 99.64 / Max: 110.85 Min: 30.18 / Avg: 194.95 / Max: 219.41 Min: 55.56 / Avg: 365.56 / Max: 407.37 Min: 32.73 / Avg: 232.73 / Max: 262.5 Min: 61.35 / Avg: 419.25 / Max: 473.7 Min: 21.47 / Avg: 157.48 / Max: 175.9 Min: 40.16 / Avg: 300.12 / Max: 342.36 Min: 75.38 / Avg: 268.45 / Max: 310.8 Min: 201.49 / Avg: 569.91 / Max: 625.34 Min: 89.32 / Avg: 273.3 / Max: 321.11 Min: 193.43 / Avg: 577.93 / Max: 636.86 Min: 61.04 / Avg: 251.84 / Max: 296.62 Min: 165.22 / Avg: 486.52 / Max: 567.17 Min: 18.63 / Avg: 215.65 / Max: 294.3 Min: 72.64 / Avg: 507.69 / Max: 575.72
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.3 Algorithm: AES-256-GCM EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 400000M 800000M 1200000M 1600000M 2000000M SE +/- 148092414.72, N = 3 SE +/- 3442274884.66, N = 3 SE +/- 818689705.82, N = 3 SE +/- 2225250504.62, N = 3 SE +/- 671752057.50, N = 3 SE +/- 1859275059.09, N = 3 SE +/- 290015345.25, N = 3 SE +/- 1791964708.17, N = 3 SE +/- 119857064.56, N = 3 SE +/- 108288170.03, N = 3 SE +/- 63026881.96, N = 3 SE +/- 112278705.55, N = 3 SE +/- 2848140790.67, N = 3 SE +/- 1397764228.85, N = 3 SE +/- 3966858436.56, N = 3 SE +/- 1320809584.03, N = 3 462309750530 415893963207 767696553667 1526505465813 859720599963 1714061761087 997539007873 1983103026700 507456728087 1011898993323 592014115453 1181148957353 700239352717 1369534328470 744387683670 1497529939377 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.3 Algorithm: AES-256-GCM EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 600M 1200M 1800M 2400M 3000M 2800094011.98 2991193842.10 2331206946.71 2286984508.79 2144285710.42 2139406659.57 2889219779.92 2844279712.45 1651813451.15 1587520126.44 1789501537.54 1745319585.66 2048623453.07 1998240264.78 2177657519.61 2193171119.61 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.84 / Avg: 165.11 / Max: 170.43 Min: 16.68 / Avg: 139.04 / Max: 145.32 Min: 29.29 / Avg: 329.31 / Max: 350.94 Min: 59.32 / Avg: 667.48 / Max: 710.25 Min: 34.11 / Avg: 400.94 / Max: 411.04 Min: 64.77 / Avg: 801.19 / Max: 822.22 Min: 22.57 / Avg: 345.26 / Max: 357.86 Min: 42.68 / Avg: 697.23 / Max: 724.85 Min: 69.15 / Avg: 307.21 / Max: 344.49 Min: 186.02 / Avg: 637.41 / Max: 707.3 Min: 76.66 / Avg: 330.83 / Max: 377.06 Min: 191.65 / Avg: 676.75 / Max: 755.44 Min: 57.11 / Avg: 341.81 / Max: 377.46 Min: 152.63 / Avg: 685.37 / Max: 754.67 Min: 23.55 / Avg: 341.83 / Max: 377.46 Min: 67.1 / Avg: 682.81 / Max: 732.1
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.3 Algorithm: AES-128-GCM EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 500000M 1000000M 1500000M 2000000M 2500000M SE +/- 530918594.60, N = 3 SE +/- 409615403.40, N = 3 SE +/- 164763342.54, N = 3 SE +/- 337702856.52, N = 3 SE +/- 570058274.06, N = 3 SE +/- 1428966819.54, N = 3 SE +/- 49279694.68, N = 3 SE +/- 1841260363.24, N = 3 SE +/- 43545680.66, N = 3 SE +/- 268465891.11, N = 3 SE +/- 56355870.99, N = 3 SE +/- 697208281.49, N = 3 SE +/- 28627701.54, N = 3 SE +/- 159598753.20, N = 3 SE +/- 1253503549.61, N = 3 SE +/- 1622921637.03, N = 3 539784931373 488965390857 897288436123 1785006879067 996007109267 1990352059967 1167221141870 2316611945283 641200848170 1277583382757 747710702660 1484669627050 768804693697 1533026300727 889567506540 1769313146013 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.3 Algorithm: AES-128-GCM EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 700M 1400M 2100M 2800M 3500M 3225126995.22 3483917754.10 2686039538.13 2638396952.00 2482510440.03 2480518885.86 3351926704.37 3291374680.15 2035473064.32 1967868118.18 2214398499.06 2168460340.43 2333237948.27 2275619248.61 2600338214.73 2589909503.87 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.13 / Avg: 167.37 / Max: 173.46 Min: 17.62 / Avg: 140.35 / Max: 146.9 Min: 28.74 / Avg: 334.06 / Max: 354.77 Min: 56.69 / Avg: 676.55 / Max: 719.01 Min: 34.83 / Avg: 401.21 / Max: 411.04 Min: 68.36 / Avg: 802.39 / Max: 822.1 Min: 22.8 / Avg: 348.22 / Max: 361.46 Min: 40.47 / Avg: 703.84 / Max: 726.54 Min: 72.06 / Avg: 315.01 / Max: 350.37 Min: 201.68 / Avg: 649.22 / Max: 717.93 Min: 73.91 / Avg: 337.66 / Max: 377.26 Min: 187.64 / Avg: 684.67 / Max: 756.52 Min: 57.48 / Avg: 329.5 / Max: 379.19 Min: 164.19 / Avg: 673.67 / Max: 754.31 Min: 21.98 / Avg: 342.1 / Max: 376.9 Min: 68.61 / Avg: 683.16 / Max: 735.37
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Road Segmentation ADAS FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 900 1800 2700 3600 4500 SE +/- 0.65, N = 3 SE +/- 0.81, N = 3 SE +/- 6.85, N = 3 SE +/- 3.90, N = 3 SE +/- 1.09, N = 3 SE +/- 5.33, N = 3 SE +/- 11.41, N = 15 SE +/- 5.40, N = 3 SE +/- 2.29, N = 3 SE +/- 1.56, N = 3 SE +/- 2.94, N = 3 SE +/- 1.74, N = 3 SE +/- 2.65, N = 3 SE +/- 0.66, N = 3 SE +/- 0.47, N = 3 SE +/- 11.53, N = 3 998.09 918.53 1743.49 3353.73 2212.41 4013.96 1671.23 4032.66 860.64 1691.79 894.87 1801.89 1068.43 1843.61 1159.66 2237.69 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130K 260K 390K 520K 650K SE +/- 314.42, N = 4 SE +/- 274.93, N = 3 SE +/- 789.76, N = 5 SE +/- 4162.49, N = 6 SE +/- 1849.65, N = 6 SE +/- 1879.23, N = 6 SE +/- 2566.89, N = 15 SE +/- 5800.63, N = 15 SE +/- 71.37, N = 3 SE +/- 441.66, N = 5 SE +/- 37.04, N = 4 SE +/- 411.20, N = 5 SE +/- 130.20, N = 4 SE +/- 398.40, N = 5 SE +/- 213.44, N = 5 SE +/- 726.00, N = 5 133284.46 130839.12 269047.28 472931.04 362511.43 551955.55 279397.45 588329.32 126171.52 246822.52 131606.86 257384.55 135125.15 261127.09 232012.77 455312.09 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6 Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 400 800 1200 1600 2000 1120.43 1212.35 1333.71 1335.18 1613.95 1398.79 1409.22 1808.53 449.46 459.15 449.20 478.19 503.44 522.54 960.89 1127.30 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.95 / Avg: 118.96 / Max: 149.72 Min: 17.84 / Avg: 107.92 / Max: 134.69 Min: 26.14 / Avg: 201.73 / Max: 306.76 Min: 56.83 / Avg: 354.21 / Max: 638.12 Min: 32.65 / Avg: 224.61 / Max: 370.12 Min: 65.04 / Avg: 394.59 / Max: 765.59 Min: 21.38 / Avg: 198.26 / Max: 322.68 Min: 40.45 / Avg: 325.31 / Max: 668.91 Min: 74.49 / Avg: 280.72 / Max: 336.16 Min: 194.99 / Avg: 537.56 / Max: 710.4 Min: 76.87 / Avg: 292.98 / Max: 353.23 Min: 196.19 / Avg: 538.25 / Max: 736.91 Min: 59.44 / Avg: 268.4 / Max: 328.81 Min: 157.28 / Avg: 499.73 / Max: 701.36 Min: 26.25 / Avg: 241.46 / Max: 364.07 Min: 67.67 / Avg: 403.9 / Max: 715.05
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Face Detection Retail FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8K 16K 24K 32K 40K SE +/- 2.45, N = 3 SE +/- 7.49, N = 3 SE +/- 2.59, N = 3 SE +/- 46.87, N = 3 SE +/- 7.70, N = 3 SE +/- 15.46, N = 3 SE +/- 8.57, N = 3 SE +/- 10.72, N = 3 SE +/- 18.13, N = 3 SE +/- 12.64, N = 3 SE +/- 7.39, N = 3 SE +/- 3.89, N = 3 SE +/- 5.45, N = 3 SE +/- 12.64, N = 3 SE +/- 4.30, N = 3 SE +/- 17.69, N = 3 9256.11 8279.63 16038.90 31974.88 17230.35 35809.59 19917.54 38546.66 9087.12 17252.27 9686.39 18561.91 9004.23 20937.58 12317.94 24676.86 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
Helsing Helsing is an open-source POSIX vampire number generator. This test profile measures the time it takes to generate vampire numbers between varying numbers of digits. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Helsing 1.0-beta Digit Range: 14 digit EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 30 60 90 120 150 SE +/- 0.27, N = 3 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.38, N = 5 SE +/- 0.14, N = 3 SE +/- 0.42, N = 3 SE +/- 0.02, N = 3 SE +/- 0.32, N = 3 SE +/- 0.53, N = 3 SE +/- 0.11, N = 3 SE +/- 0.12, N = 3 SE +/- 0.42, N = 3 SE +/- 0.26, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 105.68 120.56 62.17 33.68 56.91 30.69 48.85 26.78 123.83 65.03 107.24 58.25 93.35 50.83 89.07 48.09 1. (CC) gcc options: -O2 -pthread CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Helsing 1.0-beta CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 150 300 450 600 750 Min: 18.11 / Avg: 166.23 / Max: 176.55 Min: 17.86 / Avg: 131.19 / Max: 137.88 Min: 28.89 / Avg: 322.19 / Max: 348.84 Min: 58.49 / Avg: 624.24 / Max: 716.4 Min: 34.66 / Avg: 378.16 / Max: 412.26 Min: 67.93 / Avg: 711.96 / Max: 830.1 Min: 21.32 / Avg: 315.34 / Max: 347.42 Min: 41.44 / Avg: 591.29 / Max: 696.59 Min: 73.73 / Avg: 338.57 / Max: 364.52 Min: 210.37 / Avg: 664.49 / Max: 742.91 Min: 87.46 / Avg: 338.48 / Max: 364.16 Min: 203.6 / Avg: 662.24 / Max: 747.53 Min: 56.9 / Avg: 336.32 / Max: 376.98 Min: 173.59 / Avg: 654.95 / Max: 756.78 Min: 32.85 / Avg: 334.81 / Max: 377.7 Min: 74.95 / Avg: 643.2 / Max: 755.25
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Vehicle Detection FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3K 6K 9K 12K 15K SE +/- 2.22, N = 3 SE +/- 2.89, N = 3 SE +/- 8.39, N = 3 SE +/- 4.21, N = 3 SE +/- 3.54, N = 3 SE +/- 4.04, N = 3 SE +/- 66.33, N = 15 SE +/- 1.24, N = 3 SE +/- 8.09, N = 3 SE +/- 4.11, N = 3 SE +/- 5.68, N = 3 SE +/- 8.70, N = 3 SE +/- 0.81, N = 3 SE +/- 11.51, N = 3 SE +/- 0.73, N = 3 SE +/- 6.25, N = 3 2943.08 2763.48 5528.15 10708.91 6017.19 11735.42 5778.51 12750.59 2935.47 5576.86 2877.32 5525.56 3325.90 5483.81 4656.92 8719.37 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OSPRay Studio Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 1.0 Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20K 40K 60K 80K 100K SE +/- 172.59, N = 3 SE +/- 346.19, N = 3 SE +/- 42.05, N = 3 SE +/- 109.43, N = 3 SE +/- 193.23, N = 3 SE +/- 14.62, N = 3 SE +/- 312.83, N = 12 SE +/- 47.34, N = 3 SE +/- 134.33, N = 3 SE +/- 37.55, N = 3 SE +/- 64.91, N = 3 SE +/- 13.86, N = 3 SE +/- 102.40, N = 3 SE +/- 64.24, N = 3 SE +/- 90.37, N = 3 SE +/- 16.05, N = 3 72051 78093 43290 20750 38645 17855 37383 16960 68973 38926 62540 36179 58026 33411 50798 24249 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay Studio 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.98 / Avg: 135.52 / Max: 167.31 Min: 17.4 / Avg: 117.68 / Max: 142.04 Min: 30.06 / Avg: 244.64 / Max: 330.87 Min: 60.75 / Avg: 559.19 / Max: 686.72 Min: 30.5 / Avg: 290.44 / Max: 410.63 Min: 69.54 / Avg: 653.11 / Max: 822.49 Min: 21.62 / Avg: 219.88 / Max: 335.09 Min: 42.88 / Avg: 522.41 / Max: 691.8 Min: 69.88 / Avg: 308.64 / Max: 375.21 Min: 213.26 / Avg: 577.01 / Max: 753.84 Min: 78.44 / Avg: 306.37 / Max: 375.68 Min: 197.25 / Avg: 571.94 / Max: 754.03 Min: 60.65 / Avg: 292.18 / Max: 377.6 Min: 168.48 / Avg: 535.88 / Max: 754.67 Min: 35.11 / Avg: 281.84 / Max: 376.97 Min: 79.71 / Avg: 606.49 / Max: 752.52
PyTorch This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 14 28 42 56 70 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 SE +/- 0.10, N = 3 SE +/- 0.23, N = 3 SE +/- 0.19, N = 3 SE +/- 0.29, N = 12 SE +/- 0.04, N = 3 SE +/- 0.22, N = 12 SE +/- 0.72, N = 3 SE +/- 0.26, N = 3 SE +/- 0.43, N = 15 SE +/- 0.34, N = 15 SE +/- 0.12, N = 3 SE +/- 0.21, N = 3 SE +/- 0.19, N = 3 SE +/- 0.30, N = 15 33.76 33.18 34.65 16.24 34.69 16.82 27.53 13.76 51.05 41.65 51.86 37.84 56.44 37.55 63.30 40.83 MIN: 31.39 / MAX: 34.3 MIN: 32.26 / MAX: 33.81 MIN: 33.72 / MAX: 35.48 MIN: 11.41 / MAX: 17.76 MIN: 33.34 / MAX: 35.97 MIN: 11.96 / MAX: 19.26 MIN: 26.63 / MAX: 28.06 MIN: 9.99 / MAX: 14.97 MIN: 40.71 / MAX: 53.89 MIN: 31.6 / MAX: 43.65 MIN: 30.55 / MAX: 55.87 MIN: 19.63 / MAX: 40.77 MIN: 48.09 / MAX: 58.6 MIN: 24.5 / MAX: 40.69 MIN: 48.94 / MAX: 65.25 MIN: 15.4 / MAX: 43.41 batches/sec Per Watt
OpenBenchmarking.org batches/sec Per Watt, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0747 0.1494 0.2241 0.2988 0.3735 0.305 0.332 0.178 0.044 0.149 0.040 0.175 0.046 0.192 0.073 0.188 0.066 0.225 0.077 0.258 0.080 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better PyTorch 2.2.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 Min: 18.44 / Avg: 110.66 / Max: 125.07 Min: 17.13 / Avg: 99.8 / Max: 111.33 Min: 30.14 / Avg: 194.55 / Max: 219.58 Min: 58.93 / Avg: 369.53 / Max: 409.87 Min: 35.86 / Avg: 232.68 / Max: 261.18 Min: 63.73 / Avg: 418.64 / Max: 475.03 Min: 21.06 / Avg: 157.61 / Max: 175.55 Min: 39.07 / Avg: 298.03 / Max: 341.96 Min: 74.44 / Avg: 265.6 / Max: 309.82 Min: 202.76 / Avg: 570.95 / Max: 625.38 Min: 70.81 / Avg: 276.02 / Max: 323.04 Min: 176.16 / Avg: 577.13 / Max: 637.06 Min: 59.74 / Avg: 250.75 / Max: 298.2 Min: 161.18 / Avg: 490.32 / Max: 568.06 Min: 34.92 / Avg: 245.26 / Max: 293.93 Min: 79.08 / Avg: 511.14 / Max: 575.41
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Handwritten English Recognition FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1400 2800 4200 5600 7000 SE +/- 3.51, N = 3 SE +/- 0.47, N = 3 SE +/- 4.84, N = 3 SE +/- 9.83, N = 3 SE +/- 4.06, N = 3 SE +/- 13.59, N = 3 SE +/- 7.27, N = 3 SE +/- 14.68, N = 3 SE +/- 3.41, N = 3 SE +/- 1.09, N = 3 SE +/- 4.55, N = 3 SE +/- 2.81, N = 3 SE +/- 1.46, N = 3 SE +/- 2.46, N = 3 SE +/- 0.73, N = 3 SE +/- 3.43, N = 3 1542.61 1396.78 2739.01 5178.29 2986.33 5766.62 3280.25 6400.94 2074.30 3998.02 2243.84 4351.59 1643.43 4744.05 2924.66 5570.63 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OSPRay Studio Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 1.0 Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20K 40K 60K 80K 100K SE +/- 29.36, N = 3 SE +/- 13.58, N = 3 SE +/- 231.95, N = 3 SE +/- 756.08, N = 15 SE +/- 89.57, N = 3 SE +/- 27.74, N = 3 SE +/- 50.97, N = 3 SE +/- 28.11, N = 3 SE +/- 41.33, N = 3 SE +/- 155.80, N = 3 SE +/- 65.97, N = 3 SE +/- 141.04, N = 3 SE +/- 58.68, N = 3 SE +/- 125.33, N = 3 SE +/- 162.52, N = 3 SE +/- 47.16, N = 3 84261 91643 50019 27943 44349 20893 42945 20071 80742 45211 73366 41971 68006 38813 59522 33391 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay Studio 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.47 / Avg: 138.95 / Max: 167.41 Min: 16.93 / Avg: 120.39 / Max: 142.49 Min: 27.71 / Avg: 254.04 / Max: 331.67 Min: 60.79 / Avg: 534.58 / Max: 678.46 Min: 34.85 / Avg: 304.67 / Max: 411.05 Min: 68.36 / Avg: 672.33 / Max: 821.45 Min: 21.87 / Avg: 229.47 / Max: 334.93 Min: 43.47 / Avg: 541.66 / Max: 691.2 Min: 69.72 / Avg: 313.73 / Max: 377.15 Min: 204.64 / Avg: 589.92 / Max: 755.25 Min: 74.4 / Avg: 312.09 / Max: 377.67 Min: 194.1 / Avg: 583.5 / Max: 754.71 Min: 59.95 / Avg: 299.76 / Max: 377.1 Min: 167.48 / Avg: 554.33 / Max: 756.36 Min: 34.02 / Avg: 290.6 / Max: 377.55 Min: 81.52 / Avg: 518.28 / Max: 753.01
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20K 40K 60K 80K 100K SE +/- 126.79, N = 6 SE +/- 207.19, N = 6 SE +/- 393.67, N = 15 SE +/- 779.47, N = 7 SE +/- 386.70, N = 9 SE +/- 661.05, N = 7 SE +/- 480.71, N = 15 SE +/- 545.82, N = 5 SE +/- 313.88, N = 7 SE +/- 545.92, N = 15 SE +/- 260.16, N = 15 SE +/- 560.41, N = 15 SE +/- 243.41, N = 12 SE +/- 178.23, N = 8 SE +/- 328.18, N = 8 SE +/- 271.54, N = 6 22229.94 21684.80 53182.77 84477.72 62071.10 88487.91 43799.78 63265.01 35116.55 65396.22 35234.06 66492.44 33811.63 58128.24 63760.45 98764.93 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6 Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90 180 270 360 450 226.89 244.61 356.27 302.72 384.09 295.11 283.84 254.62 157.12 153.93 155.78 154.97 156.61 145.15 401.38 305.56 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.31 / Avg: 97.98 / Max: 151 Min: 16.9 / Avg: 88.65 / Max: 135.53 Min: 26.59 / Avg: 149.28 / Max: 317.86 Min: 55.55 / Avg: 279.06 / Max: 655.46 Min: 17.06 / Avg: 161.61 / Max: 378.41 Min: 58.39 / Avg: 299.85 / Max: 772.2 Min: 19.49 / Avg: 154.31 / Max: 321.83 Min: 38.61 / Avg: 248.47 / Max: 653.54 Min: 74.58 / Avg: 223.5 / Max: 376.44 Min: 187.39 / Avg: 424.86 / Max: 754.4 Min: 71.23 / Avg: 226.18 / Max: 377.45 Min: 189.85 / Avg: 429.08 / Max: 754.24 Min: 57.02 / Avg: 215.9 / Max: 374.98 Min: 146.08 / Avg: 400.48 / Max: 754.34 Min: 25.74 / Avg: 158.85 / Max: 377.04 Min: 54.43 / Avg: 323.22 / Max: 766.2
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3K 6K 9K 12K 15K SE +/- 9.85, N = 4 SE +/- 14.39, N = 4 SE +/- 17.53, N = 5 SE +/- 76.70, N = 10 SE +/- 29.95, N = 5 SE +/- 119.22, N = 6 SE +/- 35.67, N = 5 SE +/- 85.04, N = 8 SE +/- 15.55, N = 4 SE +/- 12.99, N = 4 SE +/- 18.08, N = 4 SE +/- 4.96, N = 4 SE +/- 6.56, N = 3 SE +/- 23.92, N = 4 SE +/- 8.92, N = 4 SE +/- 155.00, N = 15 2746.20 2718.94 5229.41 10139.69 6346.15 12296.27 5702.13 9771.41 3050.16 3816.70 3076.66 3819.83 2865.72 3960.28 4059.70 7437.81 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6 Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8 16 24 32 40 25.260 27.818 29.667 32.063 32.045 35.302 36.454 31.780 11.909 7.333 11.707 7.258 11.814 8.360 18.030 18.332 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 19.1 / Avg: 108.72 / Max: 150.47 Min: 9.26 / Avg: 97.74 / Max: 132.42 Min: 25.79 / Avg: 176.27 / Max: 304.26 Min: 54.94 / Avg: 316.24 / Max: 616.93 Min: 32.82 / Avg: 198.04 / Max: 362.78 Min: 66.48 / Avg: 348.31 / Max: 715.01 Min: 20.68 / Avg: 156.42 / Max: 284.24 Min: 41.36 / Avg: 307.47 / Max: 647.96 Min: 75.18 / Avg: 256.12 / Max: 347.73 Min: 196.72 / Avg: 520.51 / Max: 703.63 Min: 75.39 / Avg: 262.81 / Max: 362.43 Min: 196.74 / Avg: 526.3 / Max: 723.61 Min: 60.58 / Avg: 242.56 / Max: 344.06 Min: 145.91 / Avg: 473.71 / Max: 694.3 Min: 25.68 / Avg: 225.16 / Max: 371.3 Min: 67.06 / Avg: 405.72 / Max: 726.27
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.7 Preset: Exhaustive EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4 8 12 16 20 SE +/- 0.0010, N = 3 SE +/- 0.0006, N = 3 SE +/- 0.0002, N = 3 SE +/- 0.0218, N = 3 SE +/- 0.0086, N = 3 SE +/- 0.0170, N = 3 SE +/- 0.0025, N = 3 SE +/- 0.1160, N = 3 SE +/- 0.0003, N = 3 SE +/- 0.0033, N = 3 SE +/- 0.0011, N = 3 SE +/- 0.0016, N = 3 SE +/- 0.0024, N = 3 SE +/- 0.0164, N = 3 SE +/- 0.0072, N = 3 SE +/- 0.0157, N = 3 3.5146 3.1993 5.6301 10.6018 6.3027 11.9107 7.5573 14.1038 3.1204 6.2126 3.6276 7.2075 4.2689 8.3701 4.6303 8.9091 1. (CXX) g++ options: -O3 -flto -pthread MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.7 Preset: Exhaustive EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0074 0.0148 0.0222 0.0296 0.037 0.023 0.025 0.021 0.023 0.020 0.021 0.029 0.033 0.010 0.011 0.012 0.012 0.014 0.015 0.016 0.017 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.31 / Avg: 151.62 / Max: 177.01 Min: 15.2 / Avg: 129.53 / Max: 151.58 Min: 29.24 / Avg: 271.69 / Max: 338.26 Min: 58.38 / Avg: 469.68 / Max: 682.7 Min: 35.44 / Avg: 322.22 / Max: 410.64 Min: 65.5 / Avg: 555.22 / Max: 820.52 Min: 22.1 / Avg: 264.41 / Max: 366.28 Min: 41.13 / Avg: 424.16 / Max: 725.4 Min: 74.57 / Avg: 315.36 / Max: 378.01 Min: 207.97 / Avg: 591.29 / Max: 755.19 Min: 82.06 / Avg: 312.54 / Max: 376.01 Min: 207.67 / Avg: 586.49 / Max: 753.26 Min: 57.84 / Avg: 300.87 / Max: 376.96 Min: 152.83 / Avg: 551.93 / Max: 753.4 Min: 36.03 / Avg: 293.73 / Max: 378.6 Min: 80.68 / Avg: 525.87 / Max: 752.74
Blender Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.1 Blend File: BMW27 - Compute: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 7 14 21 28 35 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 4 SE +/- 0.02, N = 5 SE +/- 0.04, N = 4 SE +/- 0.03, N = 6 SE +/- 0.01, N = 4 SE +/- 0.02, N = 6 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.24, N = 3 SE +/- 0.08, N = 4 SE +/- 0.15, N = 3 SE +/- 0.08, N = 4 SE +/- 0.04, N = 3 SE +/- 0.03, N = 4 27.12 30.07 15.96 8.61 14.29 7.75 12.86 7.17 31.66 16.55 28.49 15.29 25.89 13.74 23.72 12.76 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Blender 4.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.13 / Avg: 146.16 / Max: 171.02 Min: 18.72 / Avg: 121.31 / Max: 138.75 Min: 27.61 / Avg: 264.26 / Max: 339.47 Min: 56.75 / Avg: 465.75 / Max: 695.98 Min: 35.39 / Avg: 314.77 / Max: 411.05 Min: 66.83 / Avg: 521.92 / Max: 820.17 Min: 21.13 / Avg: 247.78 / Max: 338.02 Min: 41.39 / Avg: 414.63 / Max: 680.91 Min: 72.63 / Avg: 316.78 / Max: 376.8 Min: 204.3 / Avg: 589.35 / Max: 754.59 Min: 80.66 / Avg: 311.66 / Max: 376.47 Min: 211.29 / Avg: 584.52 / Max: 755.18 Min: 57.12 / Avg: 303.66 / Max: 376.88 Min: 171.9 / Avg: 556.95 / Max: 754.1 Min: 33.51 / Avg: 297.61 / Max: 376.94 Min: 73.16 / Avg: 522.14 / Max: 732.74
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.1 Blend File: Fishy Cat - Compute: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 10 20 30 40 50 SE +/- 0.23, N = 3 SE +/- 0.07, N = 3 SE +/- 0.11, N = 3 SE +/- 0.05, N = 4 SE +/- 0.07, N = 3 SE +/- 0.02, N = 5 SE +/- 0.13, N = 3 SE +/- 0.04, N = 5 SE +/- 0.17, N = 3 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 SE +/- 0.18, N = 3 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 34.35 38.04 20.15 11.25 18.36 10.16 16.58 9.71 42.78 23.17 38.48 21.45 34.83 19.13 31.91 17.29 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Blender 4.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.84 / Avg: 152.85 / Max: 176.56 Min: 16.69 / Avg: 126.35 / Max: 143.05 Min: 30.77 / Avg: 277.68 / Max: 346.38 Min: 60.34 / Avg: 491.93 / Max: 710.82 Min: 36.44 / Avg: 322.13 / Max: 410.23 Min: 67.56 / Avg: 561.28 / Max: 821.67 Min: 22.97 / Avg: 261.77 / Max: 349.11 Min: 42.38 / Avg: 447.74 / Max: 705.75 Min: 74.34 / Avg: 321.04 / Max: 375.24 Min: 205.55 / Avg: 607.36 / Max: 752.78 Min: 74.14 / Avg: 319.1 / Max: 376.73 Min: 211.08 / Avg: 602.46 / Max: 751.81 Min: 59.73 / Avg: 313.33 / Max: 377.56 Min: 169.36 / Avg: 570.45 / Max: 755.96 Min: 23.46 / Avg: 303.18 / Max: 371.62 Min: 76.09 / Avg: 546.98 / Max: 735.92
miniBUDE MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.93, N = 3 SE +/- 0.08, N = 3 SE +/- 1.71, N = 3 SE +/- 2.28, N = 3 SE +/- 0.01, N = 3 SE +/- 3.85, N = 3 SE +/- 0.31, N = 3 SE +/- 1.48, N = 3 SE +/- 0.96, N = 4 SE +/- 0.93, N = 3 SE +/- 0.31, N = 3 SE +/- 0.99, N = 3 SE +/- 0.03, N = 3 SE +/- 1.93, N = 15 107.72 97.98 171.77 291.28 187.68 309.22 232.75 288.54 70.21 127.85 78.11 133.99 86.97 140.83 108.12 191.82 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm Billion Interactions/s Per Watt
OpenBenchmarking.org Billion Interactions/s Per Watt, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.1906 0.3812 0.5718 0.7624 0.953 0.746 0.789 0.605 0.588 0.551 0.532 0.847 0.669 0.245 0.211 0.248 0.215 0.275 0.244 0.334 0.319 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better miniBUDE 20210901 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.74 / Avg: 144.45 / Max: 156.43 Min: 16.52 / Avg: 124.23 / Max: 135.64 Min: 28.02 / Avg: 283.82 / Max: 328.33 Min: 59.15 / Avg: 495.6 / Max: 654.49 Min: 33.63 / Avg: 340.93 / Max: 409.44 Min: 68.45 / Avg: 581.3 / Max: 795.09 Min: 20.19 / Avg: 274.94 / Max: 327 Min: 42.31 / Avg: 431.39 / Max: 620.63 Min: 72.43 / Avg: 286.83 / Max: 306.63 Min: 193.86 / Avg: 605.49 / Max: 719.24 Min: 75.23 / Avg: 314.81 / Max: 352.37 Min: 203.67 / Avg: 621.91 / Max: 746.15 Min: 62.05 / Avg: 316.19 / Max: 372.81 Min: 173.4 / Avg: 578.13 / Max: 755.42 Min: 23.61 / Avg: 323.61 / Max: 376.85 Min: 63.99 / Avg: 600.52 / Max: 754.84
OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1700 3400 5100 6800 8500 SE +/- 0.13, N = 3 SE +/- 0.10, N = 3 SE +/- 23.15, N = 3 SE +/- 2.03, N = 3 SE +/- 42.85, N = 3 SE +/- 57.10, N = 3 SE +/- 0.20, N = 3 SE +/- 96.25, N = 3 SE +/- 7.81, N = 3 SE +/- 36.96, N = 3 SE +/- 23.93, N = 4 SE +/- 23.14, N = 3 SE +/- 7.77, N = 3 SE +/- 24.63, N = 3 SE +/- 0.70, N = 3 SE +/- 48.20, N = 15 2692.87 2449.45 4294.28 7281.96 4692.11 7730.47 5818.74 7213.43 1755.33 3196.12 1952.61 3349.76 2174.19 3520.71 2703.09 4795.46 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Person Vehicle Bike Detection FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2K 4K 6K 8K 10K SE +/- 2.97, N = 3 SE +/- 3.87, N = 3 SE +/- 25.68, N = 3 SE +/- 6.56, N = 3 SE +/- 18.02, N = 3 SE +/- 10.88, N = 3 SE +/- 54.09, N = 3 SE +/- 4.91, N = 3 SE +/- 7.17, N = 3 SE +/- 4.05, N = 3 SE +/- 9.37, N = 3 SE +/- 3.66, N = 3 SE +/- 4.52, N = 3 SE +/- 8.98, N = 3 SE +/- 4.97, N = 3 SE +/- 3.37, N = 3 2830.12 2506.10 5024.32 9203.19 5608.07 9752.89 5594.81 10182.16 3802.61 7127.28 4093.73 7689.32 3606.12 7862.58 5725.03 10934.17 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.7 Preset: Very Thorough EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.0007, N = 3 SE +/- 0.0012, N = 3 SE +/- 0.0006, N = 3 SE +/- 0.1965, N = 3 SE +/- 0.0254, N = 3 SE +/- 0.1284, N = 3 SE +/- 0.0006, N = 3 SE +/- 0.1961, N = 15 SE +/- 0.0016, N = 3 SE +/- 0.0055, N = 3 SE +/- 0.0016, N = 3 SE +/- 0.0022, N = 3 SE +/- 0.0054, N = 3 SE +/- 0.0379, N = 3 SE +/- 0.0294, N = 3 SE +/- 0.0007, N = 3 5.7420 5.2406 9.1844 17.0098 10.3224 19.4052 12.3187 22.0023 5.1342 10.2051 5.9702 11.8875 7.0885 13.7518 7.6240 14.6334 1. (CXX) g++ options: -O3 -flto -pthread MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.7 Preset: Very Thorough EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0117 0.0234 0.0351 0.0468 0.0585 0.038 0.041 0.034 0.036 0.032 0.036 0.047 0.052 0.016 0.017 0.019 0.020 0.023 0.025 0.026 0.028 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.13 / Avg: 151.45 / Max: 177.27 Min: 15.84 / Avg: 128.4 / Max: 151.19 Min: 25.96 / Avg: 270.94 / Max: 337.98 Min: 58.13 / Avg: 474.96 / Max: 673.2 Min: 33 / Avg: 322.42 / Max: 410.63 Min: 64.98 / Avg: 538.44 / Max: 820.03 Min: 21.13 / Avg: 262.17 / Max: 365.71 Min: 39.19 / Avg: 421.17 / Max: 704.37 Min: 74.02 / Avg: 315.87 / Max: 377.36 Min: 205.14 / Avg: 590.21 / Max: 754.13 Min: 87.45 / Avg: 313.77 / Max: 375.54 Min: 200.66 / Avg: 582.09 / Max: 758.49 Min: 61.31 / Avg: 301.81 / Max: 376.57 Min: 154.33 / Avg: 550 / Max: 755.26 Min: 35.19 / Avg: 294.16 / Max: 378.5 Min: 71.14 / Avg: 526.63 / Max: 753.88
ACES DGEMM This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOP/s, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20 40 60 80 100 SE +/- 0.16, N = 5 SE +/- 0.10, N = 5 SE +/- 0.22, N = 7 SE +/- 0.10, N = 9 SE +/- 0.40, N = 7 SE +/- 0.39, N = 9 SE +/- 0.17, N = 7 SE +/- 0.19, N = 10 SE +/- 0.04, N = 4 SE +/- 0.06, N = 7 SE +/- 0.09, N = 5 SE +/- 0.04, N = 7 SE +/- 0.09, N = 5 SE +/- 0.28, N = 7 SE +/- 0.08, N = 6 SE +/- 0.11, N = 8 21.82 21.00 39.09 68.87 45.25 78.50 44.97 84.69 19.77 39.35 22.54 44.67 23.54 45.97 29.14 58.25 1. (CC) gcc options: -O3 -march=native -fopenmp GFLOP/s Per Watt
OpenBenchmarking.org GFLOP/s Per Watt, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0657 0.1314 0.1971 0.2628 0.3285 0.198 0.220 0.214 0.228 0.196 0.215 0.252 0.292 0.071 0.078 0.081 0.090 0.093 0.101 0.121 0.149 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better ACES DGEMM 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.88 / Avg: 109.94 / Max: 158.28 Min: 16.53 / Avg: 95.47 / Max: 135.35 Min: 28.2 / Avg: 182.5 / Max: 306.04 Min: 56.61 / Avg: 302.67 / Max: 601.18 Min: 31.49 / Avg: 230.52 / Max: 409.55 Min: 63.45 / Avg: 364.55 / Max: 774.44 Min: 22.22 / Avg: 178.2 / Max: 342.37 Min: 40.76 / Avg: 290.43 / Max: 650.39 Min: 75.8 / Avg: 276.9 / Max: 361.7 Min: 193.63 / Avg: 507 / Max: 747.89 Min: 73.75 / Avg: 277.15 / Max: 377.23 Min: 194.98 / Avg: 496.59 / Max: 754.16 Min: 53.98 / Avg: 252.19 / Max: 377.52 Min: 164.24 / Avg: 453.94 / Max: 741.45 Min: 23.09 / Avg: 239.87 / Max: 363.28 Min: 68.08 / Avg: 390.95 / Max: 737.84
OSPRay Studio Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 1.0 Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 500 1000 1500 2000 2500 SE +/- 0.58, N = 3 SE +/- 1.20, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 1.15, N = 3 SE +/- 1.20, N = 3 SE +/- 1.20, N = 3 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 SE +/- 1.76, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 SE +/- 0.88, N = 3 2043 2240 1180 661 1037 561 954 531 1988 1049 1792 961 1657 894 1441 759 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay Studio 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.36 / Avg: 131.28 / Max: 167.17 Min: 16.85 / Avg: 112.81 / Max: 142.09 Min: 29.72 / Avg: 259.52 / Max: 331.33 Min: 57.88 / Avg: 533.99 / Max: 677.83 Min: 34.89 / Avg: 319.28 / Max: 410.78 Min: 62.5 / Avg: 640.62 / Max: 822.3 Min: 22.93 / Avg: 248.39 / Max: 335.61 Min: 41.9 / Avg: 518.09 / Max: 690.69 Min: 74.38 / Avg: 304.87 / Max: 378.48 Min: 203.09 / Avg: 603.55 / Max: 753.72 Min: 75.94 / Avg: 303.61 / Max: 375.87 Min: 198.09 / Avg: 602.84 / Max: 752.6 Min: 60.62 / Avg: 292.35 / Max: 377.02 Min: 171.28 / Avg: 584.71 / Max: 751.6 Min: 35.06 / Avg: 288.94 / Max: 377.45 Min: 81.36 / Avg: 571.36 / Max: 749.47
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 1.0 Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 600 1200 1800 2400 3000 SE +/- 1.00, N = 3 SE +/- 3.06, N = 3 SE +/- 1.00, N = 3 SE +/- 0.58, N = 3 SE +/- 1.53, N = 3 SE +/- 1.45, N = 3 SE +/- 0.88, N = 3 SE +/- 0.67, N = 3 SE +/- 1.20, N = 3 SE +/- 2.65, N = 3 SE +/- 1.20, N = 3 SE +/- 0.33, N = 3 SE +/- 0.88, N = 3 SE +/- 0.88, N = 3 SE +/- 1.45, N = 3 SE +/- 0.00, N = 3 2421 2649 1391 780 1225 660 1130 630 2357 1237 2123 1143 1977 1063 1709 898 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay Studio 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.61 / Avg: 131.5 / Max: 167.12 Min: 17.17 / Avg: 114.16 / Max: 142.74 Min: 31.26 / Avg: 263.32 / Max: 331.82 Min: 59.99 / Avg: 534.76 / Max: 676.92 Min: 30.75 / Avg: 323.21 / Max: 411.04 Min: 67.19 / Avg: 642.74 / Max: 821.11 Min: 22.69 / Avg: 249.28 / Max: 335.5 Min: 42.39 / Avg: 515 / Max: 688.22 Min: 71.66 / Avg: 307.39 / Max: 377.12 Min: 204.32 / Avg: 605.57 / Max: 755.38 Min: 74.64 / Avg: 306.2 / Max: 376.48 Min: 208.56 / Avg: 606.68 / Max: 753.95 Min: 61.31 / Avg: 293.93 / Max: 377.6 Min: 179.12 / Avg: 588.21 / Max: 753.37 Min: 26.01 / Avg: 290.77 / Max: 377.76 Min: 79.09 / Avg: 574.5 / Max: 751.34
LULESH LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 15K 30K 45K 60K 75K SE +/- 372.63, N = 12 SE +/- 373.78, N = 12 SE +/- 372.12, N = 3 SE +/- 349.39, N = 15 SE +/- 194.65, N = 15 SE +/- 675.87, N = 3 SE +/- 244.17, N = 3 SE +/- 427.75, N = 3 SE +/- 211.12, N = 6 SE +/- 206.20, N = 5 SE +/- 217.67, N = 6 SE +/- 207.54, N = 5 SE +/- 182.38, N = 15 SE +/- 272.44, N = 4 SE +/- 437.51, N = 4 SE +/- 359.29, N = 3 16949.59 16855.89 26082.71 48760.82 27287.26 54678.49 24218.83 46126.08 23848.54 52102.67 25831.84 53269.47 23072.30 43724.06 38433.97 70826.73 1. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi z/s Per Watt
OpenBenchmarking.org z/s Per Watt, More Is Better LULESH 2.0.3 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 149.38 162.52 143.77 123.49 130.21 122.54 134.50 122.60 100.41 97.72 103.89 100.20 105.60 88.38 155.84 143.95 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better LULESH 2.0.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 19.22 / Avg: 113.47 / Max: 157.81 Min: 17.36 / Avg: 103.72 / Max: 135.92 Min: 27.74 / Avg: 181.42 / Max: 285.77 Min: 55.4 / Avg: 394.86 / Max: 598.6 Min: 31.6 / Avg: 209.57 / Max: 345.61 Min: 68.15 / Avg: 446.2 / Max: 721.49 Min: 21.79 / Avg: 180.07 / Max: 311.6 Min: 41.83 / Avg: 376.23 / Max: 647.63 Min: 73.71 / Avg: 237.51 / Max: 333.91 Min: 198.01 / Avg: 533.2 / Max: 722.21 Min: 73.96 / Avg: 248.64 / Max: 365.61 Min: 194.87 / Avg: 531.65 / Max: 728.1 Min: 50.79 / Avg: 218.49 / Max: 329.56 Min: 165.01 / Avg: 494.73 / Max: 717.17 Min: 22.84 / Avg: 246.63 / Max: 365.09 Min: 69.2 / Avg: 492.01 / Max: 729.97
Speedb Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better Speedb 2.7 Test: Read While Writing EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8M 16M 24M 32M 40M SE +/- 88013.69, N = 15 SE +/- 86788.79, N = 3 SE +/- 170909.18, N = 3 SE +/- 746932.76, N = 15 SE +/- 118870.12, N = 3 SE +/- 651901.79, N = 15 SE +/- 172052.60, N = 3 SE +/- 803844.71, N = 15 SE +/- 88257.34, N = 3 SE +/- 171807.74, N = 3 SE +/- 93731.24, N = 7 SE +/- 197596.27, N = 3 SE +/- 62510.72, N = 3 SE +/- 239363.92, N = 3 SE +/- 143570.67, N = 4 SE +/- 195912.00, N = 15 12053901 11882103 18024993 30324596 19595777 33307574 22829426 36593204 8795408 13859254 9988176 15354400 10695360 17297027 12168413 17585737 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Speedb 2.7 Test: Read While Writing EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20K 40K 60K 80K 100K 79851.05 97163.65 63472.87 61034.97 57032.77 57884.56 93506.84 85339.03 26509.91 21183.20 29984.58 23023.23 32356.04 26260.05 37383.41 30648.42 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Speedb 2.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.32 / Avg: 150.95 / Max: 163.43 Min: 17.9 / Avg: 122.29 / Max: 139.21 Min: 26.3 / Avg: 283.98 / Max: 305.25 Min: 56.42 / Avg: 496.84 / Max: 614.4 Min: 32.81 / Avg: 343.59 / Max: 411.18 Min: 64.91 / Avg: 575.41 / Max: 820.95 Min: 22.4 / Avg: 244.15 / Max: 314.31 Min: 41.53 / Avg: 428.8 / Max: 640.33 Min: 78.68 / Avg: 331.78 / Max: 377.28 Min: 210.71 / Avg: 654.26 / Max: 754.28 Min: 74.62 / Avg: 333.11 / Max: 375.39 Min: 210.91 / Avg: 666.91 / Max: 751.13 Min: 52.35 / Avg: 330.55 / Max: 376.19 Min: 164.82 / Avg: 658.68 / Max: 750.75 Min: 34.03 / Avg: 325.5 / Max: 377.13 Min: 78.24 / Avg: 573.79 / Max: 750.97
PyTorch This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-152 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 SE +/- 0.13, N = 9 SE +/- 0.20, N = 3 SE +/- 0.13, N = 9 SE +/- 0.15, N = 3 SE +/- 0.12, N = 9 SE +/- 0.15, N = 3 SE +/- 0.09, N = 3 SE +/- 0.15, N = 3 SE +/- 0.18, N = 4 SE +/- 0.08, N = 3 SE +/- 0.13, N = 7 SE +/- 0.29, N = 3 SE +/- 0.16, N = 3 13.26 13.18 13.95 7.14 14.08 7.02 11.01 5.71 19.20 16.30 19.58 14.92 20.25 14.08 23.33 16.00 MIN: 12.98 / MAX: 13.46 MIN: 12.94 / MAX: 13.34 MIN: 13.45 / MAX: 14.29 MIN: 3.67 / MAX: 7.91 MIN: 13.43 / MAX: 14.62 MIN: 3.92 / MAX: 8.02 MIN: 10.7 / MAX: 11.43 MIN: 3.18 / MAX: 6.47 MIN: 10.93 / MAX: 20 MIN: 11.86 / MAX: 16.95 MIN: 13.13 / MAX: 20.27 MIN: 8.28 / MAX: 15.85 MIN: 17.03 / MAX: 21.21 MIN: 6.98 / MAX: 15.46 MIN: 19.78 / MAX: 24.54 MIN: 11.43 / MAX: 16.66 batches/sec Per Watt
OpenBenchmarking.org batches/sec Per Watt, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-152 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0284 0.0568 0.0852 0.1136 0.142 0.114 0.126 0.068 0.018 0.057 0.015 0.067 0.018 0.068 0.027 0.068 0.025 0.076 0.027 0.088 0.030 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better PyTorch 2.2.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 Min: 20.38 / Avg: 115.97 / Max: 125.68 Min: 16.06 / Avg: 104.56 / Max: 112.26 Min: 27.5 / Avg: 205.23 / Max: 220.66 Min: 56.85 / Avg: 395.96 / Max: 433.04 Min: 34.62 / Avg: 245.31 / Max: 264.74 Min: 63.41 / Avg: 457.61 / Max: 501.36 Min: 21.41 / Avg: 165.14 / Max: 176.25 Min: 40.42 / Avg: 321.8 / Max: 358.12 Min: 77.22 / Avg: 282.52 / Max: 309.36 Min: 199.23 / Avg: 598.06 / Max: 629.15 Min: 76.6 / Avg: 288.21 / Max: 321.16 Min: 199.11 / Avg: 604.22 / Max: 644.89 Min: 59.27 / Avg: 266.18 / Max: 295.86 Min: 161.77 / Avg: 525.94 / Max: 589.73 Min: 33.57 / Avg: 265.19 / Max: 292.3 Min: 78.85 / Avg: 540.82 / Max: 576.97
Result
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: ResNet-152 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 6 12 18 24 30 SE +/- 0.11, N = 3 SE +/- 0.09, N = 3 SE +/- 0.18, N = 3 SE +/- 0.14, N = 9 SE +/- 0.12, N = 3 SE +/- 0.13, N = 9 SE +/- 0.02, N = 3 SE +/- 0.11, N = 9 SE +/- 0.18, N = 7 SE +/- 0.09, N = 3 SE +/- 0.18, N = 7 SE +/- 0.16, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.16, N = 3 SE +/- 0.18, N = 3 13.28 13.28 14.05 7.20 13.95 7.07 10.91 5.75 19.47 16.68 19.29 14.74 20.46 14.59 23.41 16.17 MIN: 12.91 / MAX: 13.58 MIN: 12.1 / MAX: 13.57 MIN: 13.72 / MAX: 14.61 MIN: 3.59 / MAX: 8.29 MIN: 13.65 / MAX: 14.33 MIN: 3.53 / MAX: 7.83 MIN: 10.79 / MAX: 11.05 MIN: 2.81 / MAX: 6.42 MIN: 11.79 / MAX: 20.53 MIN: 7.02 / MAX: 17.07 MIN: 10.66 / MAX: 20.07 MIN: 10.63 / MAX: 15.36 MIN: 17.25 / MAX: 21.16 MIN: 10.11 / MAX: 16.06 MIN: 22.41 / MAX: 24.3 MIN: 11.54 / MAX: 16.69 batches/sec Per Watt
OpenBenchmarking.org batches/sec Per Watt, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: ResNet-152 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0286 0.0572 0.0858 0.1144 0.143 0.114 0.127 0.068 0.018 0.057 0.016 0.066 0.018 0.069 0.028 0.066 0.024 0.077 0.029 0.088 0.030 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better PyTorch 2.2.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 Min: 20.79 / Avg: 116.41 / Max: 125.98 Min: 17.18 / Avg: 104.85 / Max: 112.82 Min: 29.11 / Avg: 205.3 / Max: 221.54 Min: 57.69 / Avg: 396.9 / Max: 434.4 Min: 35.28 / Avg: 245.06 / Max: 264.04 Min: 66.98 / Avg: 454.4 / Max: 498.76 Min: 22.02 / Avg: 164.54 / Max: 175.34 Min: 40.81 / Avg: 321.68 / Max: 356.16 Min: 74.15 / Avg: 282.4 / Max: 310.35 Min: 204.68 / Avg: 599.73 / Max: 629.45 Min: 70.87 / Avg: 291.39 / Max: 322.93 Min: 191.23 / Avg: 602.51 / Max: 639.79 Min: 60.34 / Avg: 265.81 / Max: 296.61 Min: 167.37 / Avg: 511.16 / Max: 589.71 Min: 33.85 / Avg: 264.62 / Max: 291.43 Min: 83.37 / Avg: 541.35 / Max: 576.05
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Machine Translation EN To DE FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20 40 60 80 100 SE +/- 0.13, N = 3 SE +/- 0.05, N = 3 SE +/- 0.77, N = 7 SE +/- 0.44, N = 3 SE +/- 0.60, N = 11 SE +/- 0.01, N = 3 SE +/- 1.05, N = 5 SE +/- 0.59, N = 3 SE +/- 0.12, N = 3 SE +/- 0.07, N = 3 SE +/- 0.34, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.16, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 100.67 109.00 91.16 48.43 83.11 43.17 102.84 56.73 28.73 29.98 38.50 41.57 62.46 37.58 26.85 28.17 MIN: 47.39 / MAX: 138.74 MIN: 53.63 / MAX: 148.46 MIN: 43.59 / MAX: 136.45 MIN: 36.56 / MAX: 119.12 MIN: 36.66 / MAX: 131.44 MIN: 34.03 / MAX: 226.77 MIN: 47.17 / MAX: 172.5 MIN: 45.46 / MAX: 220.87 MIN: 19.52 / MAX: 209.26 MIN: 20.48 / MAX: 222.7 MIN: 25.18 / MAX: 161.13 MIN: 28.11 / MAX: 155.06 MIN: 46.12 / MAX: 79.06 MIN: 20.58 / MAX: 347.67 MIN: 20.18 / MAX: 307.38 MIN: 20.12 / MAX: 261.31 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.55 / Avg: 159.63 / Max: 174.23 Min: 17.99 / Avg: 135.1 / Max: 150.13 Min: 29.13 / Avg: 323.9 / Max: 362.47 Min: 52.92 / Avg: 645.83 / Max: 711.68 Min: 31.03 / Avg: 374.75 / Max: 410.94 Min: 69.87 / Avg: 750.94 / Max: 819.6 Min: 20.94 / Avg: 299.63 / Max: 344.93 Min: 39.45 / Avg: 607.43 / Max: 679.21 Min: 70.61 / Avg: 331.05 / Max: 366.72 Min: 196.83 / Avg: 663.81 / Max: 748.85 Min: 74.79 / Avg: 330.98 / Max: 373.36 Min: 203.57 / Avg: 662.79 / Max: 751.8 Min: 59.14 / Avg: 321.95 / Max: 371.05 Min: 167.2 / Avg: 656.03 / Max: 736.45 Min: 34.43 / Avg: 325.81 / Max: 362.26 Min: 86.44 / Avg: 651.59 / Max: 748.25
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50K 100K 150K 200K 250K SE +/- 167.65, N = 8 SE +/- 301.54, N = 8 SE +/- 1153.71, N = 15 SE +/- 2707.72, N = 14 SE +/- 1372.89, N = 15 SE +/- 7207.31, N = 15 SE +/- 884.47, N = 9 SE +/- 1776.37, N = 11 SE +/- 436.97, N = 10 SE +/- 1415.96, N = 9 SE +/- 382.59, N = 10 SE +/- 1298.67, N = 9 SE +/- 152.92, N = 9 SE +/- 1262.04, N = 9 SE +/- 406.29, N = 9 SE +/- 1422.07, N = 7 63930.64 62449.55 116725.04 194942.32 134994.14 244738.81 125567.07 245447.72 85910.95 172614.10 86407.46 171436.47 83448.70 163333.21 107321.89 220893.78 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6 Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300 600 900 1200 1500 906.46 953.72 1070.68 823.46 1119.20 928.22 1265.55 988.65 499.08 476.05 480.22 454.47 517.49 496.64 858.24 757.48 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.11 / Avg: 70.53 / Max: 141.22 Min: 16.49 / Avg: 65.48 / Max: 127.89 Min: 27.57 / Avg: 109.02 / Max: 286.49 Min: 55.9 / Avg: 236.74 / Max: 614.68 Min: 29.67 / Avg: 120.62 / Max: 340.18 Min: 62.46 / Avg: 263.66 / Max: 710.66 Min: 21.58 / Avg: 99.22 / Max: 284.09 Min: 40.32 / Avg: 248.27 / Max: 531.63 Min: 72.77 / Avg: 172.14 / Max: 347.7 Min: 183.3 / Avg: 362.6 / Max: 714.29 Min: 73.21 / Avg: 179.93 / Max: 359.94 Min: 193.92 / Avg: 377.22 / Max: 729.73 Min: 59.14 / Avg: 161.26 / Max: 333.63 Min: 142.29 / Avg: 328.87 / Max: 698.64 Min: 25.53 / Avg: 125.05 / Max: 346.36 Min: 70.18 / Avg: 291.62 / Max: 590.65
Algebraic Multi-Grid Benchmark AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1000M 2000M 3000M 4000M 5000M SE +/- 2902028.33, N = 3 SE +/- 291866.98, N = 3 SE +/- 1066998.75, N = 3 SE +/- 8808919.44, N = 3 SE +/- 4963892.66, N = 3 SE +/- 6340323.98, N = 3 SE +/- 433397.56, N = 3 SE +/- 235140242.43, N = 12 SE +/- 3359086.30, N = 3 SE +/- 7233468.75, N = 3 SE +/- 1778769.08, N = 3 SE +/- 9164253.67, N = 3 SE +/- 468935.32, N = 3 SE +/- 2438223.49, N = 3 SE +/- 4000850.45, N = 3 SE +/- 9014927.20, N = 3 1150052667 1147108000 2282802000 4324930333 2376149000 4467281333 2275352667 2783039083 1588496667 3156261667 1597993333 3205161000 1605097000 3124402667 1814760000 3609646333 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi Figure Of Merit Per Watt
OpenBenchmarking.org Figure Of Merit Per Watt, More Is Better Algebraic Multi-Grid Benchmark 1.2 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2M 4M 6M 8M 10M 9344903.38 10387087.22 9442032.07 8578578.70 8293777.78 7422966.05 10610537.82 5533330.98 5380716.03 5320019.08 5363427.41 5337195.71 5415151.06 5278683.30 6349445.21 6520382.60 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Algebraic Multi-Grid Benchmark 1.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.02 / Avg: 123.07 / Max: 150.22 Min: 16.12 / Avg: 110.44 / Max: 131.86 Min: 23.58 / Avg: 241.77 / Max: 304.95 Min: 58.67 / Avg: 504.15 / Max: 640.85 Min: 27.09 / Avg: 286.5 / Max: 375.22 Min: 63.8 / Avg: 601.82 / Max: 781.06 Min: 20 / Avg: 214.44 / Max: 294.78 Min: 42.41 / Avg: 502.96 / Max: 635.91 Min: 75.43 / Avg: 295.22 / Max: 373.47 Min: 189.38 / Avg: 593.28 / Max: 741.95 Min: 59.54 / Avg: 297.94 / Max: 371.37 Min: 186.85 / Avg: 600.53 / Max: 731.72 Min: 58.01 / Avg: 296.41 / Max: 368.28 Min: 164.81 / Avg: 591.89 / Max: 729.05 Min: 25.82 / Avg: 285.81 / Max: 370.88 Min: 67.59 / Avg: 553.59 / Max: 733.88
srsRAN Project srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mbps, More Is Better srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Total EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2K 4K 6K 8K 10K SE +/- 25.25, N = 3 SE +/- 19.02, N = 13 SE +/- 0.20, N = 3 SE +/- 74.95, N = 4 SE +/- 55.17, N = 3 SE +/- 2.29, N = 3 SE +/- 63.64, N = 5 SE +/- 76.45, N = 3 SE +/- 0.09, N = 3 SE +/- 26.47, N = 3 SE +/- 18.30, N = 3 SE +/- 0.15, N = 3 SE +/- 18.43, N = 3 SE +/- 47.68, N = 6 SE +/- 18.09, N = 3 SE +/- 0.12, N = 3 3193.0 2818.5 5239.6 6835.7 5656.9 6999.7 5973.8 8396.7 2170.7 3787.0 2514.0 4386.6 2616.9 4853.6 3088.7 5493.4 MIN: 1941.2 / MAX: 3219 MIN: 1696.8 / MAX: 2893.2 MIN: 3593.7 / MAX: 5240 MIN: 6107.2 / MAX: 6950.8 MIN: 3595.4 / MAX: 5767.2 MIN: 6424.8 / MAX: 7003.9 MIN: 4286.9 / MAX: 6228.4 MIN: 8064.1 / MAX: 8549.4 MIN: 1388.2 / MAX: 2170.8 MIN: 2545.7 / MAX: 3839.9 MIN: 1550.4 / MAX: 2532.5 MIN: 2970.2 / MAX: 4386.9 MIN: 1591.7 / MAX: 2653.8 MIN: 3181.8 / MAX: 4929.4 MIN: 1852.6 / MAX: 3124.9 MIN: 3703.4 / MAX: 5493.6 1. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl Mbps Per Watt
OpenBenchmarking.org Mbps Per Watt, More Is Better srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Total EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 6 12 18 24 30 24.996 26.846 20.666 12.702 18.052 10.725 23.406 16.197 7.971 6.643 8.854 7.552 10.299 8.808 13.247 10.976 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.17 / Avg: 127.74 / Max: 148.29 Min: 14.51 / Avg: 104.99 / Max: 125.17 Min: 27.31 / Avg: 253.54 / Max: 296.82 Min: 60.3 / Avg: 538.17 / Max: 621.29 Min: 31.36 / Avg: 313.37 / Max: 378.12 Min: 67.45 / Avg: 652.65 / Max: 788.02 Min: 21.5 / Avg: 255.23 / Max: 297.3 Min: 37.54 / Avg: 518.41 / Max: 623.41 Min: 74.32 / Avg: 272.33 / Max: 304.79 Min: 206.4 / Avg: 570.06 / Max: 634.71 Min: 73.76 / Avg: 283.68 / Max: 318.07 Min: 196.65 / Avg: 580.82 / Max: 660.22 Min: 62.91 / Avg: 254.1 / Max: 294.51 Min: 156.44 / Avg: 551.02 / Max: 636.6 Min: 25.89 / Avg: 233.17 / Max: 276.77 Min: 82.2 / Avg: 500.5 / Max: 584.22
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Machine Translation EN To DE FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200 400 600 800 1000 SE +/- 0.39, N = 3 SE +/- 0.13, N = 3 SE +/- 4.59, N = 7 SE +/- 8.96, N = 3 SE +/- 4.29, N = 11 SE +/- 0.37, N = 3 SE +/- 6.34, N = 5 SE +/- 11.78, N = 3 SE +/- 1.21, N = 3 SE +/- 1.99, N = 3 SE +/- 4.09, N = 3 SE +/- 1.13, N = 3 SE +/- 0.17, N = 3 SE +/- 3.48, N = 3 SE +/- 0.54, N = 3 SE +/- 0.85, N = 3 317.38 293.07 525.88 990.16 576.98 1110.49 621.42 1127.05 415.96 795.63 467.08 889.18 477.77 793.64 592.20 1128.78 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
Aircrack-ng Aircrack-ng is a tool for assessing WiFi/WLAN network security. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org k/s, More Is Better Aircrack-ng 1.7 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40K 80K 120K 160K 200K SE +/- 338.71, N = 3 SE +/- 319.60, N = 3 SE +/- 207.11, N = 3 SE +/- 1622.56, N = 3 SE +/- 714.86, N = 3 SE +/- 36.42, N = 3 SE +/- 1027.07, N = 15 SE +/- 122.31, N = 3 SE +/- 1131.21, N = 15 SE +/- 1953.08, N = 3 SE +/- 1197.41, N = 15 SE +/- 840.59, N = 3 SE +/- 773.70, N = 3 153818.03 141831.78 186675.43 206619.29 177989.89 117588.99 83665.76 134048.96 65663.82 147469.33 53914.93 153584.20 64181.29 1. (CXX) g++ options: -std=gnu++17 -O3 -fvisibility=hidden -fcommon -rdynamic -lnl-3 -lnl-genl-3 -lpcre -lsqlite3 -lpthread -lz -lssl -lcrypto -lhwloc -ldl -lm -pthread k/s Per Watt
OpenBenchmarking.org k/s Per Watt, More Is Better Aircrack-ng 1.7 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300 600 900 1200 1500 1057.15 1173.57 702.97 585.84 763.99 377.96 142.86 433.27 112.13 479.36 97.52 507.73 128.68 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Aircrack-ng 1.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 17.23 / Avg: 145.5 / Max: 168.72 Min: 15.46 / Avg: 120.86 / Max: 145.49 Min: 29.99 / Avg: 265.55 / Max: 307.89 Min: 34.08 / Avg: 352.69 / Max: 410.53 Min: 10.09 / Avg: 232.97 / Max: 273.01 Min: 73.18 / Avg: 311.12 / Max: 377.22 Min: 195.2 / Avg: 585.65 / Max: 659.01 Min: 77.17 / Avg: 309.39 / Max: 375.45 Min: 193.67 / Avg: 585.6 / Max: 670.08 Min: 55.34 / Avg: 307.64 / Max: 375.71 Min: 160.46 / Avg: 552.88 / Max: 630.35 Min: 25.19 / Avg: 302.49 / Max: 376.59 Min: 78.32 / Avg: 498.75 / Max: 586.29
EPYC 9654 2P: The test run did not produce a result.
EPYC 9684X 2P: The test run did not produce a result.
EPYC 9754 2P: The test run did not produce a result.
Blender Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.1 Blend File: Junkshop - Compute: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 10 20 30 40 50 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 4 SE +/- 0.13, N = 3 SE +/- 0.08, N = 4 SE +/- 0.08, N = 3 SE +/- 0.02, N = 4 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.19, N = 3 SE +/- 0.24, N = 3 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 SE +/- 0.09, N = 3 35.82 38.86 21.02 12.87 18.65 11.47 17.70 11.21 42.85 24.99 39.17 23.45 35.96 21.15 32.08 18.31 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Blender 4.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.72 / Avg: 148.28 / Max: 170.5 Min: 18.29 / Avg: 127.16 / Max: 145.93 Min: 16.01 / Avg: 272.23 / Max: 337.39 Min: 59.75 / Avg: 483.45 / Max: 681.8 Min: 35.12 / Avg: 316.57 / Max: 411.05 Min: 63.79 / Avg: 553.98 / Max: 822.05 Min: 22.09 / Avg: 255.66 / Max: 340.71 Min: 39.86 / Avg: 438.61 / Max: 691.32 Min: 75.79 / Avg: 322.02 / Max: 374.53 Min: 206.83 / Avg: 603.68 / Max: 753.22 Min: 85.92 / Avg: 320.19 / Max: 376.39 Min: 208.58 / Avg: 605.47 / Max: 751.55 Min: 59.14 / Avg: 311.52 / Max: 376.21 Min: 169.62 / Avg: 575.47 / Max: 754.28 Min: 24.83 / Avg: 303.23 / Max: 378.56 Min: 77.61 / Avg: 545.55 / Max: 722.02
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.1 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 14 28 42 56 70 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.14, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.16, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 18.47 16.92 30.61 52.67 34.64 60.37 37.74 63.85 19.91 35.93 22.36 39.34 24.64 42.82 27.10 48.76 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 3.1 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.032 0.064 0.096 0.128 0.16 0.135 0.142 0.112 0.091 0.094 0.084 0.130 0.111 0.063 0.055 0.070 0.060 0.079 0.068 0.088 0.077 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay 3.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.46 / Avg: 136.83 / Max: 158.83 Min: 16.8 / Avg: 119.14 / Max: 138.68 Min: 29.59 / Avg: 273.92 / Max: 319.11 Min: 60.47 / Avg: 577.89 / Max: 656.09 Min: 35.77 / Avg: 367.12 / Max: 410.2 Min: 65.72 / Avg: 718.49 / Max: 821.14 Min: 23.6 / Avg: 289.58 / Max: 332.39 Min: 42.2 / Avg: 573.28 / Max: 670.75 Min: 76.02 / Avg: 317.76 / Max: 373.98 Min: 206.3 / Avg: 652.41 / Max: 753.05 Min: 78.8 / Avg: 318.09 / Max: 375.14 Min: 207.37 / Avg: 650.92 / Max: 752.3 Min: 61.9 / Avg: 311.6 / Max: 375.6 Min: 174.41 / Avg: 631.92 / Max: 708.59 Min: 26.13 / Avg: 308.08 / Max: 376.91 Min: 77.1 / Avg: 630.9 / Max: 750.84
WRF WRF, the Weather Research and Forecasting Model, is a "next-generation mesoscale numerical weather prediction system designed for both atmospheric research and operational forecasting applications. It features two dynamical cores, a data assimilation system, and a software architecture supporting parallel computation and system extensibility." Learn more via the OpenBenchmarking.org test page.
miniBUDE MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 0.04, N = 6 SE +/- 0.03, N = 6 SE +/- 1.18, N = 7 SE +/- 2.97, N = 13 SE +/- 3.27, N = 13 SE +/- 6.15, N = 12 SE +/- 0.70, N = 9 SE +/- 3.98, N = 15 SE +/- 0.12, N = 4 SE +/- 1.28, N = 12 SE +/- 0.84, N = 15 SE +/- 1.85, N = 12 SE +/- 0.61, N = 5 SE +/- 0.81, N = 7 SE +/- 0.20, N = 6 SE +/- 1.02, N = 8 107.35 98.21 136.95 217.85 151.95 236.61 229.91 214.62 63.62 108.40 70.84 114.90 82.85 133.63 106.40 199.73 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm Billion Interactions/s Per Watt
OpenBenchmarking.org Billion Interactions/s Per Watt, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.3179 0.6358 0.9537 1.2716 1.5895 1.008 1.043 0.706 0.655 0.640 0.611 1.413 0.732 0.262 0.228 0.280 0.231 0.324 0.291 0.445 0.503 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better miniBUDE 20210901 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.42 / Avg: 106.48 / Max: 155.79 Min: 17.11 / Avg: 94.19 / Max: 135.5 Min: 26.83 / Avg: 194.1 / Max: 318.58 Min: 53.82 / Avg: 332.37 / Max: 634.66 Min: 29.7 / Avg: 237.41 / Max: 409.16 Min: 64.39 / Avg: 387.38 / Max: 798.84 Min: 19.01 / Avg: 162.7 / Max: 323.16 Min: 39.64 / Avg: 293.27 / Max: 587.35 Min: 71.42 / Avg: 242.48 / Max: 294.63 Min: 179.1 / Avg: 476.05 / Max: 713.49 Min: 72.35 / Avg: 253.17 / Max: 324.27 Min: 188.58 / Avg: 496.58 / Max: 737.62 Min: 60.74 / Avg: 255.91 / Max: 343.2 Min: 149.89 / Avg: 459.2 / Max: 669.79 Min: 21.92 / Avg: 239.11 / Max: 377.96 Min: 63.44 / Avg: 396.77 / Max: 754.62
OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1300 2600 3900 5200 6500 SE +/- 0.97, N = 6 SE +/- 0.82, N = 6 SE +/- 29.52, N = 7 SE +/- 74.36, N = 13 SE +/- 81.83, N = 13 SE +/- 153.85, N = 12 SE +/- 17.54, N = 9 SE +/- 99.42, N = 15 SE +/- 2.91, N = 4 SE +/- 32.01, N = 12 SE +/- 21.10, N = 15 SE +/- 46.19, N = 12 SE +/- 15.23, N = 5 SE +/- 20.15, N = 7 SE +/- 5.07, N = 6 SE +/- 25.44, N = 8 2683.83 2455.16 3423.63 5446.16 3798.70 5915.25 5747.82 5365.42 1590.50 2709.93 1771.07 2872.56 2071.19 3340.69 2659.88 4993.34 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Crown EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 0.04, N = 4 SE +/- 0.07, N = 4 SE +/- 0.08, N = 6 SE +/- 0.43, N = 7 SE +/- 0.20, N = 7 SE +/- 1.23, N = 8 SE +/- 0.64, N = 6 SE +/- 1.04, N = 7 SE +/- 0.05, N = 4 SE +/- 0.14, N = 6 SE +/- 0.02, N = 4 SE +/- 0.48, N = 6 SE +/- 0.07, N = 5 SE +/- 0.12, N = 6 SE +/- 0.05, N = 5 SE +/- 0.07, N = 7 62.19 58.28 110.53 176.21 130.79 214.02 126.15 190.08 58.42 107.63 65.33 116.61 70.05 124.71 82.97 153.16 MIN: 61.2 / MAX: 63.88 MIN: 57.28 / MAX: 60.13 MIN: 107.72 / MAX: 115.26 MIN: 163.69 / MAX: 195.97 MIN: 126.46 / MAX: 138.83 MIN: 181.82 / MAX: 248.2 MIN: 122.27 / MAX: 133.25 MIN: 164.86 / MAX: 219.06 MIN: 55.81 / MAX: 61.43 MIN: 102.22 / MAX: 114.8 MIN: 60.89 / MAX: 69.91 MIN: 107.47 / MAX: 129.72 MIN: 63.26 / MAX: 79.13 MIN: 115.3 / MAX: 143.82 MIN: 79.09 / MAX: 87.11 MIN: 143.78 / MAX: 166.08 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Crown EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.1415 0.283 0.4245 0.566 0.7075 0.482 0.517 0.499 0.468 0.522 0.499 0.629 0.532 0.206 0.205 0.232 0.226 0.258 0.252 0.331 0.363 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Embree 4.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.65 / Avg: 128.92 / Max: 171.09 Min: 16.58 / Avg: 112.82 / Max: 146.36 Min: 27.85 / Avg: 221.71 / Max: 341.98 Min: 58.37 / Avg: 376.52 / Max: 669.54 Min: 32.54 / Avg: 250.47 / Max: 410.99 Min: 60.99 / Avg: 428.94 / Max: 819.87 Min: 20.38 / Avg: 200.39 / Max: 335.38 Min: 39.31 / Avg: 357.16 / Max: 678.39 Min: 71.98 / Avg: 284.3 / Max: 376.72 Min: 192.27 / Avg: 525.15 / Max: 751.36 Min: 75.4 / Avg: 281.67 / Max: 377.64 Min: 178.12 / Avg: 516.97 / Max: 755.2 Min: 60.79 / Avg: 271.03 / Max: 377.88 Min: 171.82 / Avg: 493.94 / Max: 734.13 Min: 26.17 / Avg: 250.53 / Max: 384.62 Min: 65.93 / Avg: 422.5 / Max: 748.67
GROMACS The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2024 Implementation: MPI CPU - Input: water_GMX50_bare EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.004, N = 3 SE +/- 0.004, N = 3 SE +/- 0.034, N = 3 SE +/- 0.189, N = 6 SE +/- 0.052, N = 3 SE +/- 0.224, N = 14 SE +/- 0.032, N = 3 SE +/- 0.067, N = 3 SE +/- 0.019, N = 3 SE +/- 0.025, N = 3 SE +/- 0.007, N = 3 SE +/- 0.032, N = 3 SE +/- 0.010, N = 3 SE +/- 0.019, N = 3 SE +/- 0.009, N = 3 SE +/- 0.046, N = 3 6.499 6.246 12.514 20.234 15.222 22.525 13.242 21.305 7.545 13.392 8.282 14.127 8.762 15.450 10.477 18.540 1. (CXX) g++ options: -O3 -lm Ns Per Day Per Watt
OpenBenchmarking.org Ns Per Day Per Watt, More Is Better GROMACS 2024 Implementation: MPI CPU - Input: water_GMX50_bare EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0119 0.0238 0.0357 0.0476 0.0595 0.047 0.052 0.048 0.040 0.048 0.037 0.053 0.047 0.024 0.022 0.027 0.023 0.029 0.026 0.036 0.035 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better GROMACS 2024 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.05 / Avg: 137.24 / Max: 164.83 Min: 16.45 / Avg: 120.24 / Max: 142.62 Min: 27.69 / Avg: 260.21 / Max: 348.03 Min: 56.1 / Avg: 505.98 / Max: 708.92 Min: 33.95 / Avg: 317.95 / Max: 409.49 Min: 65.5 / Avg: 601.18 / Max: 820.89 Min: 22.18 / Avg: 250.1 / Max: 348.03 Min: 40.72 / Avg: 455.51 / Max: 701.31 Min: 76.31 / Avg: 314 / Max: 374.89 Min: 201.3 / Avg: 602.79 / Max: 753.69 Min: 75.59 / Avg: 311.94 / Max: 375.77 Min: 205.45 / Avg: 603.46 / Max: 753.72 Min: 54.7 / Avg: 306.77 / Max: 376.19 Min: 166.32 / Avg: 584.52 / Max: 753.32 Min: 24.97 / Avg: 289.21 / Max: 376.32 Min: 75.68 / Avg: 535.3 / Max: 753.28
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Slow EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 14 28 42 56 70 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 4 SE +/- 0.50, N = 5 SE +/- 0.07, N = 4 SE +/- 0.26, N = 5 SE +/- 0.05, N = 4 SE +/- 0.47, N = 8 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 4 23.71 22.46 38.90 61.07 40.00 62.82 38.47 54.88 17.52 26.00 18.24 28.96 20.23 31.24 24.23 40.92 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Slow EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0513 0.1026 0.1539 0.2052 0.2565 0.199 0.228 0.169 0.149 0.146 0.135 0.202 0.169 0.063 0.048 0.064 0.052 0.077 0.063 0.087 0.080 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 17.7 / Avg: 119.03 / Max: 142.44 Min: 15.1 / Avg: 98.52 / Max: 118.03 Min: 28.16 / Avg: 229.8 / Max: 294.04 Min: 58.04 / Avg: 409.96 / Max: 602.7 Min: 30.18 / Avg: 273.17 / Max: 355.5 Min: 59.79 / Avg: 464.35 / Max: 680.95 Min: 21.44 / Avg: 190.72 / Max: 249.47 Min: 38.94 / Avg: 325.22 / Max: 485.91 Min: 80.03 / Avg: 279.05 / Max: 314.96 Min: 210.38 / Avg: 537.83 / Max: 613.5 Min: 67.53 / Avg: 286.96 / Max: 322.9 Min: 195.26 / Avg: 552.42 / Max: 652.44 Min: 58.28 / Avg: 262.77 / Max: 303.74 Min: 168.67 / Avg: 496.87 / Max: 590.71 Min: 28.15 / Avg: 278.75 / Max: 351.38 Min: 71.05 / Avg: 508.41 / Max: 687.83
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.17 Model: yolov4 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 6 12 18 24 30 SE +/- 0.10588, N = 4 SE +/- 0.09781, N = 5 SE +/- 0.08554, N = 8 SE +/- 0.04207, N = 3 SE +/- 0.12519, N = 3 SE +/- 0.08454, N = 3 SE +/- 0.05314, N = 3 SE +/- 0.09575, N = 3 SE +/- 0.16821, N = 15 SE +/- 0.29451, N = 15 SE +/- 0.15263, N = 15 SE +/- 0.09595, N = 14 SE +/- 0.04802, N = 3 SE +/- 0.06825, N = 3 SE +/- 0.05084, N = 3 SE +/- 0.09991, N = 3 8.87446 8.75233 9.86018 6.41229 9.73685 7.24869 8.29533 6.74977 18.21790 14.48310 16.91660 14.45830 19.85930 15.55550 22.96890 16.96720 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: yolov4 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 30 60 90 120 150 SE +/- 1.36, N = 4 SE +/- 1.32, N = 5 SE +/- 0.88, N = 8 SE +/- 1.03, N = 3 SE +/- 1.34, N = 3 SE +/- 1.60, N = 3 SE +/- 0.77, N = 3 SE +/- 2.08, N = 3 SE +/- 0.55, N = 15 SE +/- 1.43, N = 15 SE +/- 0.55, N = 15 SE +/- 0.47, N = 14 SE +/- 0.12, N = 3 SE +/- 0.28, N = 3 SE +/- 0.10, N = 3 SE +/- 0.35, N = 3 112.73 114.31 101.47 155.96 102.73 137.99 120.56 148.21 54.96 69.45 59.18 69.20 50.35 64.29 43.54 58.94 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Llamafile Mozilla's Llamafile allows distributing and running large language models (LLMs) as a single file. Llamafile aims to make open-source LLMs more accessible to developers and users. Llamafile supports a variety of models, CPUs and GPUs, and other options. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.7 Test: llava-v1.5-7b-q4 - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 6 12 18 24 30 SE +/- 0.11, N = 4 SE +/- 0.07, N = 4 SE +/- 0.02, N = 5 SE +/- 0.07, N = 4 SE +/- 0.10, N = 5 SE +/- 0.02, N = 4 SE +/- 0.05, N = 4 SE +/- 0.08, N = 4 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.15, N = 4 SE +/- 0.09, N = 15 SE +/- 0.17, N = 4 SE +/- 1.08, N = 12 21.00 20.65 25.75 14.47 25.38 14.29 20.64 10.94 7.67 7.77 8.13 7.26 20.95 10.44 16.61 7.49 Tokens Per Second Per Watt
OpenBenchmarking.org Tokens Per Second Per Watt, More Is Better Llamafile 0.7 Test: llava-v1.5-7b-q4 - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0502 0.1004 0.1506 0.2008 0.251 0.202 0.223 0.138 0.037 0.110 0.031 0.126 0.032 0.027 0.014 0.028 0.013 0.080 0.019 0.068 0.015 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Llamafile 0.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.78 / Avg: 104.22 / Max: 157.47 Min: 14.6 / Avg: 92.71 / Max: 131.37 Min: 28.61 / Avg: 186.7 / Max: 318.2 Min: 57.8 / Avg: 388.15 / Max: 621.67 Min: 32.28 / Avg: 231.13 / Max: 402.04 Min: 62.86 / Avg: 462.55 / Max: 790.45 Min: 20.15 / Avg: 163.65 / Max: 315.47 Min: 40.53 / Avg: 343.52 / Max: 619.26 Min: 72.53 / Avg: 283.07 / Max: 370.83 Min: 205.99 / Avg: 559.87 / Max: 716.53 Min: 72.54 / Avg: 288.22 / Max: 376.39 Min: 191.45 / Avg: 566.01 / Max: 704.67 Min: 56.03 / Avg: 262.55 / Max: 376.09 Min: 152.91 / Avg: 537.58 / Max: 745.57 Min: 34.18 / Avg: 244.07 / Max: 354.9 Min: 81.53 / Avg: 512.88 / Max: 707.9
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.1 Benchmark: particle_volume/ao/real_time EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 13 26 39 52 65 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 17.97 16.45 28.10 47.61 32.52 55.34 35.25 56.69 16.05 29.50 18.55 33.11 21.93 38.22 30.84 54.12 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 3.1 Benchmark: particle_volume/ao/real_time EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.029 0.058 0.087 0.116 0.145 0.121 0.129 0.096 0.079 0.085 0.072 0.114 0.095 0.060 0.043 0.056 0.049 0.065 0.056 0.091 0.080 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay 3.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.75 / Avg: 148.12 / Max: 155.49 Min: 16.71 / Avg: 127.12 / Max: 134.69 Min: 28.62 / Avg: 291.85 / Max: 303.75 Min: 60.35 / Avg: 600.41 / Max: 624.73 Min: 36.05 / Avg: 383.1 / Max: 399.03 Min: 61.93 / Avg: 764.28 / Max: 797.27 Min: 23.11 / Avg: 308.84 / Max: 321.58 Min: 42.49 / Avg: 596.49 / Max: 627.65 Min: 74.5 / Avg: 268.43 / Max: 329.84 Min: 200.34 / Avg: 682.67 / Max: 705.07 Min: 78.47 / Avg: 334.1 / Max: 367.59 Min: 195.3 / Avg: 681.19 / Max: 725.09 Min: 59.83 / Avg: 336.31 / Max: 366.79 Min: 176.11 / Avg: 681.46 / Max: 714.09 Min: 35 / Avg: 338.48 / Max: 364.64 Min: 77.58 / Avg: 677.36 / Max: 715.65
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.1 Benchmark: particle_volume/scivis/real_time EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 13 26 39 52 65 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 17.94 16.45 28.07 47.53 32.52 55.26 35.15 56.45 16.09 29.35 18.49 33.13 21.89 38.23 30.68 53.93 Items Per Second Per Watt
OpenBenchmarking.org Items Per Second Per Watt, More Is Better OSPRay 3.1 Benchmark: particle_volume/scivis/real_time EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0326 0.0652 0.0978 0.1304 0.163 0.139 0.145 0.107 0.086 0.098 0.081 0.133 0.107 0.063 0.047 0.061 0.053 0.074 0.061 0.106 0.091 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OSPRay 3.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.42 / Avg: 129.01 / Max: 155.58 Min: 16.2 / Avg: 113.27 / Max: 134.96 Min: 28.04 / Avg: 261.28 / Max: 302.78 Min: 60.52 / Avg: 553.25 / Max: 625.16 Min: 35.05 / Avg: 332.91 / Max: 399.84 Min: 67.89 / Avg: 682.63 / Max: 796.5 Min: 22.25 / Avg: 264.01 / Max: 319.18 Min: 42.57 / Avg: 526.9 / Max: 623.56 Min: 74.87 / Avg: 255.3 / Max: 321.66 Min: 191.04 / Avg: 627.62 / Max: 711.24 Min: 76.7 / Avg: 301.86 / Max: 366.25 Min: 200.54 / Avg: 630.06 / Max: 725.99 Min: 59.59 / Avg: 296.81 / Max: 366.58 Min: 168.54 / Avg: 625.45 / Max: 728.13 Min: 25.32 / Avg: 288.49 / Max: 368.86 Min: 67.98 / Avg: 591.05 / Max: 725.39
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 14 28 42 56 70 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 4 SE +/- 0.56, N = 5 SE +/- 0.01, N = 4 SE +/- 0.40, N = 5 SE +/- 0.04, N = 4 SE +/- 0.52, N = 15 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.27, N = 4 24.31 23.02 39.58 61.11 40.50 63.48 39.08 54.57 18.13 26.69 18.84 29.62 20.97 32.04 24.80 41.58 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0522 0.1044 0.1566 0.2088 0.261 0.200 0.232 0.181 0.151 0.148 0.135 0.207 0.169 0.065 0.050 0.066 0.054 0.080 0.065 0.090 0.082 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 17.56 / Avg: 121.69 / Max: 146.12 Min: 15.7 / Avg: 99.33 / Max: 124.48 Min: 26.12 / Avg: 219.12 / Max: 294.65 Min: 56.07 / Avg: 404.28 / Max: 599.6 Min: 34.18 / Avg: 273.99 / Max: 354.7 Min: 66.41 / Avg: 469.54 / Max: 685.47 Min: 20.71 / Avg: 188.76 / Max: 248.4 Min: 38.26 / Avg: 323.31 / Max: 486.48 Min: 77.93 / Avg: 280.57 / Max: 315.37 Min: 209.31 / Avg: 533.47 / Max: 617.43 Min: 69.51 / Avg: 286.2 / Max: 325.03 Min: 189.77 / Avg: 550.48 / Max: 652.57 Min: 61.73 / Avg: 263.01 / Max: 306.52 Min: 168.08 / Avg: 490.42 / Max: 593.69 Min: 33.8 / Avg: 276.47 / Max: 362.01 Min: 75.75 / Avg: 509.74 / Max: 682.24
OpenVKL OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Items / Sec, More Is Better OpenVKL 2.0.0 Benchmark: vklBenchmarkCPU ISPC EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 900 1800 2700 3600 4500 SE +/- 1.20, N = 3 SE +/- 8.17, N = 3 SE +/- 1.20, N = 3 SE +/- 8.67, N = 3 SE +/- 1.00, N = 3 SE +/- 5.36, N = 3 SE +/- 0.88, N = 3 SE +/- 26.33, N = 3 SE +/- 6.66, N = 3 SE +/- 8.62, N = 3 SE +/- 13.96, N = 3 SE +/- 4.36, N = 3 SE +/- 1.53, N = 3 SE +/- 6.74, N = 3 SE +/- 2.08, N = 3 SE +/- 13.62, N = 3 1266 1177 2077 3335 2396 3906 2508 4012 1243 1776 1356 1852 1734 2716 2004 3222 MIN: 98 / MAX: 15312 MIN: 91 / MAX: 14403 MIN: 160 / MAX: 26161 MIN: 299 / MAX: 40960 MIN: 187 / MAX: 29055 MIN: 350 / MAX: 44654 MIN: 207 / MAX: 30640 MIN: 382 / MAX: 47557 MIN: 95 / MAX: 19286 MIN: 177 / MAX: 27371 MIN: 107 / MAX: 20909 MIN: 191 / MAX: 28947 MIN: 104 / MAX: 26188 MIN: 188 / MAX: 30888 MIN: 134 / MAX: 29750 MIN: 248 / MAX: 36519 Items / Sec Per Watt
OpenBenchmarking.org Items / Sec Per Watt, More Is Better OpenVKL 2.0.0 Benchmark: vklBenchmarkCPU ISPC EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 9.163 9.780 7.601 6.102 6.861 5.763 9.457 7.673 3.798 2.745 4.129 2.853 5.391 4.287 6.261 5.106 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVKL 2.0.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.98 / Avg: 138.16 / Max: 167.46 Min: 18.05 / Avg: 120.35 / Max: 144.9 Min: 28.49 / Avg: 273.27 / Max: 332.24 Min: 57.11 / Avg: 546.55 / Max: 678.78 Min: 33.91 / Avg: 349.21 / Max: 411.78 Min: 67.14 / Avg: 677.76 / Max: 821.93 Min: 21.69 / Avg: 265.2 / Max: 336.16 Min: 40.48 / Avg: 522.89 / Max: 678.27 Min: 75.39 / Avg: 327.25 / Max: 376.15 Min: 209.11 / Avg: 647.03 / Max: 756.03 Min: 78.05 / Avg: 328.39 / Max: 378.11 Min: 190.07 / Avg: 649.08 / Max: 754.84 Min: 57.62 / Avg: 321.66 / Max: 378.49 Min: 172.02 / Avg: 633.5 / Max: 755.3 Min: 35.93 / Avg: 320.1 / Max: 378.34 Min: 79.05 / Avg: 630.97 / Max: 751.76
Xcompact3d Incompact3d Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: X3D-benchmarking input.i3d EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 SE +/- 0.92, N = 3 SE +/- 0.22, N = 3 SE +/- 4.73, N = 3 SE +/- 3.26, N = 9 SE +/- 3.06, N = 9 SE +/- 2.20, N = 12 SE +/- 5.39, N = 3 SE +/- 9.13, N = 9 SE +/- 1.25, N = 3 SE +/- 0.04, N = 3 SE +/- 0.14, N = 3 SE +/- 0.01, N = 3 SE +/- 3.88, N = 4 SE +/- 0.05, N = 3 SE +/- 2.51, N = 3 SE +/- 0.37, N = 3 567.29 568.25 466.55 248.87 309.85 167.13 534.06 253.32 407.59 189.69 351.64 187.22 357.01 196.81 323.25 171.38 1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.39 / Avg: 140.74 / Max: 150 Min: 16.95 / Avg: 129.67 / Max: 136.25 Min: 30.46 / Avg: 296.94 / Max: 321.61 Min: 56.03 / Avg: 638.9 / Max: 677.63 Min: 33.87 / Avg: 359.91 / Max: 399.74 Min: 66.18 / Avg: 750.96 / Max: 804.8 Min: 11.29 / Avg: 271.32 / Max: 306.5 Min: 38.21 / Avg: 591.17 / Max: 651.55 Min: 73.01 / Avg: 342.5 / Max: 356.35 Min: 203.67 / Avg: 685.84 / Max: 731.99 Min: 80.13 / Avg: 345.35 / Max: 367.23 Min: 201.23 / Avg: 686.09 / Max: 746.29 Min: 59.93 / Avg: 344.75 / Max: 358.18 Min: 158.32 / Avg: 681.46 / Max: 722.98 Min: 23.89 / Avg: 333.9 / Max: 356.36 Min: 77.71 / Avg: 675.14 / Max: 722.78
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Person Vehicle Bike Detection FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.11, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 11.22 12.67 9.47 5.20 8.48 4.91 11.35 6.27 12.54 13.34 13.59 14.42 16.50 15.04 11.17 11.69 MIN: 6.69 / MAX: 20.05 MIN: 7.38 / MAX: 24.53 MIN: 5.96 / MAX: 23.01 MIN: 4.52 / MAX: 23.03 MIN: 5 / MAX: 22.91 MIN: 4.44 / MAX: 21.92 MIN: 6.34 / MAX: 32.69 MIN: 5.66 / MAX: 25.66 MIN: 9.17 / MAX: 21.01 MIN: 9.67 / MAX: 39.84 MIN: 9.53 / MAX: 21.61 MIN: 9.74 / MAX: 43.95 MIN: 11.95 / MAX: 28.87 MIN: 12.32 / MAX: 32.87 MIN: 10.18 / MAX: 19.36 MIN: 10.28 / MAX: 42.12 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.03 / Avg: 158.07 / Max: 171.2 Min: 15.14 / Avg: 128.27 / Max: 140.82 Min: 29.5 / Avg: 315.59 / Max: 344.89 Min: 35.7 / Avg: 634.64 / Max: 688.41 Min: 34.6 / Avg: 368.08 / Max: 403.09 Min: 70.55 / Avg: 710.38 / Max: 794.96 Min: 22.15 / Avg: 293.51 / Max: 329.03 Min: 43.98 / Avg: 604.36 / Max: 672.48 Min: 73.58 / Avg: 327.03 / Max: 377.73 Min: 206.04 / Avg: 658.8 / Max: 752.52 Min: 83.78 / Avg: 329.87 / Max: 374.74 Min: 204.78 / Avg: 655.73 / Max: 753.37 Min: 60.78 / Avg: 328.2 / Max: 380.56 Min: 174.42 / Avg: 651.12 / Max: 753.13 Min: 33.49 / Avg: 325.49 / Max: 379.56 Min: 84.22 / Avg: 639.31 / Max: 759.19
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 32 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2000M 4000M 6000M 8000M 10000M SE +/- 1501110.70, N = 3 SE +/- 1928730.15, N = 3 SE +/- 1419311.26, N = 3 SE +/- 46730658.51, N = 3 SE +/- 11426625.83, N = 3 SE +/- 34610274.65, N = 3 SE +/- 3768731.00, N = 3 SE +/- 2936551.27, N = 3 SE +/- 1320353.49, N = 3 SE +/- 2447674.63, N = 3 SE +/- 1338323.99, N = 3 SE +/- 895048.11, N = 3 SE +/- 4021608.30, N = 3 SE +/- 10496401.50, N = 3 SE +/- 2516611.48, N = 3 SE +/- 6711433.03, N = 3 3062200000 2789600000 5271933333 7513333333 5862466667 7935266667 6588200000 6439000000 2381800000 4670966667 2769066667 5367033333 3243800000 6116566667 3167200000 6077300000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 32 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5M 10M 15M 20M 25M 21992105.60 23304528.15 17844515.21 13431732.69 16171511.91 12478169.67 22727990.09 16713460.67 8576477.66 7263532.04 9061082.03 8335236.32 10249944.70 9661394.05 10180344.31 9752693.41 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.82 / Avg: 139.24 / Max: 159.66 Min: 16.68 / Avg: 119.7 / Max: 139.58 Min: 30.54 / Avg: 295.44 / Max: 336.72 Min: 59.09 / Avg: 559.37 / Max: 644.46 Min: 34.34 / Avg: 362.52 / Max: 411.45 Min: 65.54 / Avg: 635.93 / Max: 742.68 Min: 21.58 / Avg: 289.87 / Max: 330.19 Min: 40.61 / Avg: 385.26 / Max: 440.96 Min: 74.86 / Avg: 277.71 / Max: 312.01 Min: 186.6 / Avg: 643.07 / Max: 754.55 Min: 78.49 / Avg: 305.6 / Max: 341.63 Min: 204.52 / Avg: 643.9 / Max: 752.56 Min: 59.61 / Avg: 316.47 / Max: 377.01 Min: 160.12 / Avg: 633.09 / Max: 754.38 Min: 25.92 / Avg: 311.11 / Max: 357.53 Min: 70.49 / Avg: 623.14 / Max: 733.57
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Obj EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 4 SE +/- 1.58, N = 15 SE +/- 0.11, N = 4 SE +/- 1.72, N = 15 SE +/- 0.07, N = 4 SE +/- 3.26, N = 12 SE +/- 0.08, N = 3 SE +/- 0.76, N = 3 SE +/- 0.09, N = 3 SE +/- 0.89, N = 15 SE +/- 0.09, N = 3 SE +/- 0.22, N = 4 SE +/- 0.35, N = 3 SE +/- 0.29, N = 4 64.33 63.34 113.28 175.60 133.48 210.44 133.93 200.16 64.96 104.84 71.83 112.72 79.38 132.28 97.19 173.22 MIN: 63.73 / MAX: 65.21 MIN: 62.81 / MAX: 64.1 MIN: 111.7 / MAX: 116.19 MIN: 154.41 / MAX: 194.22 MIN: 130.86 / MAX: 138.79 MIN: 168.12 / MAX: 235.13 MIN: 131.91 / MAX: 136.58 MIN: 148.48 / MAX: 223.22 MIN: 63.12 / MAX: 67.81 MIN: 101.05 / MAX: 112.02 MIN: 69.55 / MAX: 76.73 MIN: 102 / MAX: 125.96 MIN: 76.07 / MAX: 85.98 MIN: 125 / MAX: 146.52 MIN: 92.96 / MAX: 102.04 MIN: 164.07 / MAX: 187.8 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Obj EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.2201 0.4402 0.6603 0.8804 1.1005 0.642 0.713 0.693 0.575 0.716 0.602 0.978 0.759 0.263 0.222 0.291 0.240 0.368 0.320 0.523 0.579 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Embree 4.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.71 / Avg: 100.16 / Max: 169.93 Min: 17.83 / Avg: 88.8 / Max: 143.53 Min: 28.7 / Avg: 163.54 / Max: 333.81 Min: 56.83 / Avg: 305.41 / Max: 649.39 Min: 32.74 / Avg: 186.36 / Max: 410.63 Min: 65.49 / Avg: 349.82 / Max: 819.75 Min: 19.95 / Avg: 136.96 / Max: 327.05 Min: 36.73 / Avg: 263.84 / Max: 656.51 Min: 73.93 / Avg: 247.01 / Max: 377.15 Min: 193.13 / Avg: 472.12 / Max: 755.04 Min: 86.41 / Avg: 246.7 / Max: 376.83 Min: 187.33 / Avg: 469.55 / Max: 757.04 Min: 61.45 / Avg: 215.69 / Max: 377.02 Min: 164.62 / Avg: 413.97 / Max: 758.06 Min: 26.36 / Avg: 185.86 / Max: 375.76 Min: 66.3 / Avg: 299.17 / Max: 755.1
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.7 Preset: Thorough EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 30 60 90 120 150 SE +/- 0.02, N = 4 SE +/- 0.02, N = 3 SE +/- 0.01, N = 4 SE +/- 0.73, N = 15 SE +/- 0.06, N = 5 SE +/- 1.46, N = 12 SE +/- 0.01, N = 5 SE +/- 1.20, N = 5 SE +/- 0.01, N = 3 SE +/- 0.05, N = 4 SE +/- 0.03, N = 4 SE +/- 0.03, N = 5 SE +/- 0.01, N = 4 SE +/- 0.19, N = 5 SE +/- 0.05, N = 4 SE +/- 0.55, N = 5 41.02 37.96 65.30 105.41 73.38 116.12 86.01 121.30 37.01 72.33 42.98 83.34 51.30 86.98 54.45 97.94 1. (CXX) g++ options: -O3 -flto -pthread MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.7 Preset: Thorough EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.1136 0.2272 0.3408 0.4544 0.568 0.359 0.384 0.338 0.319 0.330 0.316 0.505 0.456 0.139 0.153 0.164 0.179 0.214 0.203 0.244 0.262 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 15.96 / Avg: 114.12 / Max: 173.87 Min: 15.09 / Avg: 98.95 / Max: 145.65 Min: 28.71 / Avg: 193.35 / Max: 335.29 Min: 55.87 / Avg: 330.17 / Max: 623.66 Min: 33.53 / Avg: 222.24 / Max: 407.18 Min: 63.05 / Avg: 367.67 / Max: 734.43 Min: 20.4 / Avg: 170.2 / Max: 353.81 Min: 38.72 / Avg: 265.99 / Max: 577.68 Min: 72.98 / Avg: 265.38 / Max: 375.06 Min: 204.94 / Avg: 472.75 / Max: 744.21 Min: 78.02 / Avg: 261.71 / Max: 374.53 Min: 195.71 / Avg: 466.49 / Max: 741.06 Min: 60.85 / Avg: 239.45 / Max: 374.18 Min: 147.85 / Avg: 428.35 / Max: 731.41 Min: 34.05 / Avg: 223.18 / Max: 369.47 Min: 76.68 / Avg: 373.98 / Max: 732.13
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Slow EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 10 20 30 40 50 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.21, N = 4 SE +/- 0.04, N = 3 SE +/- 0.12, N = 4 SE +/- 0.04, N = 3 SE +/- 0.23, N = 4 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 17.04 15.77 28.01 40.96 29.18 43.39 29.00 37.13 13.36 19.75 13.89 22.14 15.71 24.71 16.06 27.11 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Slow EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0331 0.0662 0.0993 0.1324 0.1655 0.130 0.147 0.108 0.086 0.092 0.080 0.127 0.099 0.044 0.034 0.045 0.038 0.054 0.045 0.055 0.049 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.85 / Avg: 130.95 / Max: 149.92 Min: 16.27 / Avg: 107.27 / Max: 124.79 Min: 28.73 / Avg: 258.29 / Max: 311.87 Min: 57.99 / Avg: 473.23 / Max: 616.87 Min: 32.45 / Avg: 317.42 / Max: 382.18 Min: 66.68 / Avg: 543.26 / Max: 734.77 Min: 21.98 / Avg: 227.72 / Max: 282.74 Min: 40.4 / Avg: 375.03 / Max: 518.77 Min: 76.03 / Avg: 301.55 / Max: 333.2 Min: 196.37 / Avg: 575.73 / Max: 648.93 Min: 38.16 / Avg: 306.83 / Max: 339.09 Min: 181.76 / Avg: 590.16 / Max: 673.18 Min: 61.29 / Avg: 289.12 / Max: 322.82 Min: 171.75 / Avg: 547.66 / Max: 641.35 Min: 26.18 / Avg: 291.45 / Max: 378.34 Min: 75.78 / Avg: 550.27 / Max: 756.12
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 512 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 500M 1000M 1500M 2000M 2500M SE +/- 776881.23, N = 3 SE +/- 170391.71, N = 3 SE +/- 520683.31, N = 3 SE +/- 11753202.87, N = 3 SE +/- 2676440.42, N = 3 SE +/- 14870701.99, N = 3 SE +/- 1300000.00, N = 3 SE +/- 1650252.51, N = 3 SE +/- 569766.03, N = 3 SE +/- 4238841.56, N = 3 SE +/- 740202.67, N = 3 SE +/- 6728628.72, N = 3 SE +/- 1722723.94, N = 3 SE +/- 5630373.98, N = 3 SE +/- 240370.09, N = 3 SE +/- 14465476.14, N = 3 821013333 741740000 1362166667 2129533333 1570400000 2394333333 1744700000 1734800000 838530000 1444066667 944550000 1518933333 1020833333 1616966667 1178766667 2027000000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 512 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1.4M 2.8M 4.2M 5.6M 7M 5998206.24 6532499.05 4853033.94 4001237.14 4387111.54 3603186.16 6413920.52 4657018.09 2666847.49 2269719.80 2982249.17 2388645.74 3267678.46 2571656.77 3805850.46 3272343.49 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.86 / Avg: 136.88 / Max: 157.55 Min: 17.93 / Avg: 113.55 / Max: 130.59 Min: 29.69 / Avg: 280.68 / Max: 320.8 Min: 53.43 / Avg: 532.22 / Max: 613.02 Min: 35.59 / Avg: 357.96 / Max: 411.45 Min: 67.21 / Avg: 664.5 / Max: 779.27 Min: 20.69 / Avg: 272.02 / Max: 314.58 Min: 40.06 / Avg: 372.51 / Max: 431.46 Min: 74.44 / Avg: 314.43 / Max: 376.72 Min: 210.59 / Avg: 636.23 / Max: 755.25 Min: 77.34 / Avg: 316.72 / Max: 376.86 Min: 198.99 / Avg: 635.9 / Max: 752.9 Min: 60.01 / Avg: 312.4 / Max: 376.93 Min: 157.59 / Avg: 628.76 / Max: 754.2 Min: 25.3 / Avg: 309.72 / Max: 376.79 Min: 70.53 / Avg: 619.43 / Max: 752.7
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 0.03, N = 5 SE +/- 0.05, N = 5 SE +/- 0.03, N = 6 SE +/- 1.54, N = 8 SE +/- 0.07, N = 7 SE +/- 2.04, N = 8 SE +/- 0.12, N = 7 SE +/- 1.88, N = 7 SE +/- 0.04, N = 5 SE +/- 0.63, N = 6 SE +/- 0.11, N = 5 SE +/- 1.20, N = 15 SE +/- 0.05, N = 5 SE +/- 0.13, N = 7 SE +/- 0.13, N = 6 SE +/- 0.26, N = 8 75.21 73.91 132.49 185.71 155.12 231.43 156.39 208.31 75.47 125.75 83.49 129.75 91.99 152.75 112.66 199.61 MIN: 74.55 / MAX: 76.96 MIN: 73.25 / MAX: 75 MIN: 130.77 / MAX: 135.13 MIN: 154.13 / MAX: 227.12 MIN: 152.39 / MAX: 161.68 MIN: 187.58 / MAX: 274.64 MIN: 153.92 / MAX: 160 MIN: 162.41 / MAX: 258.13 MIN: 72.93 / MAX: 78.77 MIN: 120.06 / MAX: 135.13 MIN: 80.58 / MAX: 88.97 MIN: 110.67 / MAX: 151.91 MIN: 88.57 / MAX: 98.54 MIN: 143.8 / MAX: 168.98 MIN: 107.88 / MAX: 117.19 MIN: 185.98 / MAX: 216.64 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.194 0.388 0.582 0.776 0.97 0.613 0.706 0.735 0.518 0.662 0.564 0.862 0.620 0.276 0.246 0.312 0.253 0.428 0.323 0.494 0.508 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Embree 4.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.74 / Avg: 122.78 / Max: 171.96 Min: 16.8 / Avg: 104.66 / Max: 145.45 Min: 28.26 / Avg: 180.38 / Max: 335.4 Min: 55.21 / Avg: 358.56 / Max: 646.17 Min: 31.26 / Avg: 234.18 / Max: 410.22 Min: 59.43 / Avg: 410.15 / Max: 817.39 Min: 20.65 / Avg: 181.41 / Max: 329.67 Min: 42.62 / Avg: 335.83 / Max: 642.99 Min: 66.39 / Avg: 273.07 / Max: 377.18 Min: 200.84 / Avg: 511.17 / Max: 752.67 Min: 74.25 / Avg: 267.87 / Max: 377.44 Min: 185.96 / Avg: 511.91 / Max: 753.56 Min: 60.02 / Avg: 215.03 / Max: 377.71 Min: 166.53 / Avg: 472.72 / Max: 754.96 Min: 24.21 / Avg: 228.28 / Max: 376.12 Min: 66.59 / Avg: 393.21 / Max: 754.66
High Performance Conjugate Gradient HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 X Y Z: 144 144 144 - RT: 60 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 15 30 45 60 75 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.28, N = 5 SE +/- 0.51, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.29, N = 3 SE +/- 0.37, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.94, N = 3 22.20 22.13 28.32 42.52 27.58 44.95 23.63 41.87 25.24 54.41 30.59 62.64 31.07 60.85 35.14 69.30 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi GFLOP/s Per Watt
OpenBenchmarking.org GFLOP/s Per Watt, More Is Better High Performance Conjugate Gradient 3.1 X Y Z: 144 144 144 - RT: 60 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.038 0.076 0.114 0.152 0.19 0.156 0.169 0.104 0.069 0.084 0.062 0.099 0.074 0.073 0.079 0.089 0.092 0.090 0.088 0.103 0.101 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better High Performance Conjugate Gradient 3.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.73 / Avg: 141.98 / Max: 160.79 Min: 18.07 / Avg: 131.23 / Max: 136.75 Min: 28.57 / Avg: 273.24 / Max: 316.28 Min: 55.46 / Avg: 612.13 / Max: 685.77 Min: 33.89 / Avg: 330.25 / Max: 391.86 Min: 67.43 / Avg: 722.2 / Max: 816.31 Min: 16.99 / Avg: 238.17 / Max: 311.47 Min: 42.67 / Avg: 563.2 / Max: 666.57 Min: 74.31 / Avg: 344.79 / Max: 363.36 Min: 192.7 / Avg: 690.6 / Max: 722.1 Min: 76.3 / Avg: 345.19 / Max: 369.22 Min: 208.84 / Avg: 683.79 / Max: 753.27 Min: 62.11 / Avg: 344.43 / Max: 357.3 Min: 162.34 / Avg: 688.79 / Max: 719.93 Min: 24.94 / Avg: 340.38 / Max: 366.08 Min: 78.99 / Avg: 685.14 / Max: 717.63
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.17 Model: GPT-2 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 60 120 180 240 300 SE +/- 0.28, N = 3 SE +/- 0.63, N = 3 SE +/- 1.37, N = 5 SE +/- 0.13, N = 3 SE +/- 0.70, N = 3 SE +/- 0.46, N = 3 SE +/- 0.15, N = 3 SE +/- 1.43, N = 15 SE +/- 2.01, N = 15 SE +/- 0.58, N = 3 SE +/- 1.88, N = 3 SE +/- 1.08, N = 3 SE +/- 1.50, N = 3 SE +/- 1.05, N = 3 SE +/- 1.19, N = 3 SE +/- 8.14, N = 15 89.21 88.57 126.82 102.57 162.09 161.95 89.74 86.69 176.33 141.95 168.86 132.71 222.03 180.03 270.33 236.65 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: GPT-2 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 SE +/- 0.03478, N = 3 SE +/- 0.07964, N = 3 SE +/- 0.08751, N = 5 SE +/- 0.01221, N = 3 SE +/- 0.02661, N = 3 SE +/- 0.01742, N = 3 SE +/- 0.01807, N = 3 SE +/- 0.20120, N = 15 SE +/- 0.06596, N = 15 SE +/- 0.02902, N = 3 SE +/- 0.06620, N = 3 SE +/- 0.06105, N = 3 SE +/- 0.03075, N = 3 SE +/- 0.03218, N = 3 SE +/- 0.01631, N = 3 SE +/- 0.15083, N = 15 11.20470 11.28640 7.88584 9.74614 6.16668 6.17190 11.13920 11.57740 5.67700 7.04111 5.91961 7.53259 4.49976 5.55101 3.69649 4.29531 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Intel Open Image Denoise Open Image Denoise is a denoising library for ray-tracing and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.2 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1.1453 2.2906 3.4359 4.5812 5.7265 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 4 SE +/- 0.01, N = 5 SE +/- 0.02, N = 4 SE +/- 0.00, N = 5 SE +/- 0.03, N = 6 SE +/- 0.01, N = 5 SE +/- 0.00, N = 4 SE +/- 0.02, N = 5 SE +/- 0.00, N = 4 SE +/- 0.02, N = 5 SE +/- 0.00, N = 4 SE +/- 0.01, N = 6 SE +/- 0.00, N = 5 SE +/- 0.03, N = 6 1.80 1.69 2.73 3.49 2.66 3.47 3.16 3.57 2.27 3.49 2.39 3.66 2.71 4.48 3.32 5.09 Images / Sec Per Watt
OpenBenchmarking.org Images / Sec Per Watt, More Is Better Intel Open Image Denoise 2.2 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0036 0.0072 0.0108 0.0144 0.018 0.015 0.016 0.012 0.008 0.009 0.007 0.015 0.010 0.008 0.006 0.008 0.007 0.010 0.009 0.013 0.012 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Intel Open Image Denoise 2.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.74 / Avg: 123.7 / Max: 154.19 Min: 17.42 / Avg: 107.91 / Max: 132.12 Min: 27.41 / Avg: 233.36 / Max: 317.91 Min: 57.66 / Avg: 424.59 / Max: 613.78 Min: 32.55 / Avg: 282.53 / Max: 386.33 Min: 63.03 / Avg: 499.92 / Max: 728.7 Min: 21.62 / Avg: 205.54 / Max: 298.26 Min: 40.05 / Avg: 372.52 / Max: 545.78 Min: 73.66 / Avg: 286.68 / Max: 372.63 Min: 208.03 / Avg: 550.26 / Max: 748.26 Min: 76.35 / Avg: 284.74 / Max: 374.26 Min: 193.7 / Avg: 538.98 / Max: 749.89 Min: 59.91 / Avg: 268.44 / Max: 367.04 Min: 165.7 / Avg: 496.66 / Max: 745.69 Min: 26.77 / Avg: 249.97 / Max: 378.66 Min: 80.74 / Avg: 440.63 / Max: 738.2
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.2 Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.54 1.08 1.62 2.16 2.7 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 4 SE +/- 0.00, N = 3 SE +/- 0.01, N = 4 0.86 0.80 1.31 1.67 1.28 1.64 1.50 1.71 1.08 1.72 1.14 1.74 1.29 2.18 1.58 2.40 Images / Sec Per Watt
OpenBenchmarking.org Images / Sec Per Watt, More Is Better Intel Open Image Denoise 2.2 Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0016 0.0032 0.0048 0.0064 0.008 0.006 0.007 0.005 0.003 0.004 0.003 0.006 0.004 0.003 0.003 0.004 0.003 0.004 0.004 0.005 0.004 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Intel Open Image Denoise 2.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.84 / Avg: 136.47 / Max: 153.36 Min: 17.12 / Avg: 118.45 / Max: 132.79 Min: 26.15 / Avg: 266.61 / Max: 316.67 Min: 60.67 / Avg: 504.98 / Max: 617.56 Min: 35.65 / Avg: 329.68 / Max: 387.49 Min: 66.51 / Avg: 585.9 / Max: 726.86 Min: 22.02 / Avg: 239.18 / Max: 293.59 Min: 40.88 / Avg: 433.56 / Max: 546.1 Min: 75.68 / Avg: 314.57 / Max: 370.97 Min: 206.4 / Avg: 605.57 / Max: 749.65 Min: 78.93 / Avg: 313 / Max: 373.31 Min: 198.59 / Avg: 608.01 / Max: 748.72 Min: 59.44 / Avg: 304.48 / Max: 364.78 Min: 169.3 / Avg: 573.61 / Max: 745.77 Min: 35.29 / Avg: 290 / Max: 378.6 Min: 79.35 / Avg: 534.18 / Max: 740.08
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Noise Suppression Poconet-Like FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 7 14 21 28 35 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.16, N = 14 SE +/- 0.08, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 SE +/- 0.45, N = 12 SE +/- 0.10, N = 3 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.00, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 17.30 18.49 16.32 23.12 13.52 22.20 22.67 28.10 11.37 17.15 12.72 18.72 13.95 15.96 9.40 13.15 MIN: 10.72 / MAX: 37.75 MIN: 11.59 / MAX: 39.13 MIN: 8.11 / MAX: 42.33 MIN: 10.62 / MAX: 67.83 MIN: 6.46 / MAX: 38.94 MIN: 9.55 / MAX: 75.9 MIN: 8.31 / MAX: 66.6 MIN: 12.27 / MAX: 100.22 MIN: 9.17 / MAX: 32.12 MIN: 10.81 / MAX: 59.34 MIN: 9.56 / MAX: 30.66 MIN: 12.15 / MAX: 64.81 MIN: 8.67 / MAX: 65.92 MIN: 10.38 / MAX: 43.64 MIN: 7.66 / MAX: 25.07 MIN: 7.28 / MAX: 47.97 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.46 / Avg: 161.25 / Max: 175.66 Min: 17.49 / Avg: 137.08 / Max: 149.27 Min: 29.75 / Avg: 315.18 / Max: 357.38 Min: 61.14 / Avg: 581.56 / Max: 658.32 Min: 34.79 / Avg: 374.52 / Max: 409.97 Min: 67.53 / Avg: 716.78 / Max: 816.61 Min: 20.78 / Avg: 292.63 / Max: 342.23 Min: 41.91 / Avg: 488.86 / Max: 570.14 Min: 73.68 / Avg: 329.15 / Max: 372.33 Min: 204.87 / Avg: 656.57 / Max: 725.7 Min: 85.86 / Avg: 330.63 / Max: 384.53 Min: 205.34 / Avg: 657.37 / Max: 743.06 Min: 60.51 / Avg: 327.19 / Max: 379.48 Min: 171.01 / Avg: 649.04 / Max: 743.64 Min: 34.54 / Avg: 324.07 / Max: 405.42 Min: 43.35 / Avg: 637.42 / Max: 766.05
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 10 20 30 40 50 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.23, N = 4 SE +/- 0.03, N = 3 SE +/- 0.46, N = 4 SE +/- 0.05, N = 3 SE +/- 0.28, N = 11 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.16, N = 3 SE +/- 0.05, N = 3 19.25 17.80 31.44 42.03 32.74 45.26 32.44 38.12 15.17 22.25 15.68 24.65 17.85 27.65 16.64 30.37 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0374 0.0748 0.1122 0.1496 0.187 0.148 0.166 0.150 0.090 0.105 0.086 0.145 0.105 0.051 0.039 0.051 0.042 0.062 0.051 0.059 0.057 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.2 / Avg: 130.37 / Max: 151.17 Min: 16.52 / Avg: 107.55 / Max: 128.61 Min: 29.07 / Avg: 209.83 / Max: 311.67 Min: 56.62 / Avg: 464.69 / Max: 608.77 Min: 34.79 / Avg: 310.62 / Max: 383.33 Min: 66.05 / Avg: 527.19 / Max: 710.22 Min: 22.31 / Avg: 224.13 / Max: 286.22 Min: 40.54 / Avg: 362.19 / Max: 493.12 Min: 76.55 / Avg: 299.38 / Max: 332.83 Min: 208.68 / Avg: 571.96 / Max: 640.22 Min: 74.3 / Avg: 306.72 / Max: 336.87 Min: 190.07 / Avg: 582.58 / Max: 673.85 Min: 60.4 / Avg: 287.16 / Max: 327.69 Min: 168.73 / Avg: 543.59 / Max: 642.52 Min: 26.08 / Avg: 283.27 / Max: 364.78 Min: 76.25 / Avg: 536.26 / Max: 744.89
Intel Open Image Denoise Open Image Denoise is a denoising library for ray-tracing and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.2 Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1.1318 2.2636 3.3954 4.5272 5.659 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 4 SE +/- 0.01, N = 5 SE +/- 0.00, N = 4 SE +/- 0.00, N = 5 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 SE +/- 0.00, N = 4 SE +/- 0.01, N = 5 SE +/- 0.00, N = 4 SE +/- 0.04, N = 5 SE +/- 0.00, N = 4 SE +/- 0.03, N = 6 SE +/- 0.00, N = 5 SE +/- 0.03, N = 6 1.79 1.69 2.73 3.48 2.64 3.45 3.12 3.59 2.26 3.51 2.38 3.65 2.71 4.41 3.31 5.03 Images / Sec Per Watt
OpenBenchmarking.org Images / Sec Per Watt, More Is Better Intel Open Image Denoise 2.2 Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0036 0.0072 0.0108 0.0144 0.018 0.014 0.016 0.012 0.008 0.009 0.007 0.015 0.010 0.008 0.006 0.008 0.007 0.010 0.009 0.013 0.011 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Intel Open Image Denoise 2.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 19.01 / Avg: 124.67 / Max: 154.19 Min: 17.3 / Avg: 107.16 / Max: 132.19 Min: 25.93 / Avg: 230.11 / Max: 317.81 Min: 56.38 / Avg: 425.53 / Max: 615.08 Min: 32.59 / Avg: 284.36 / Max: 386.47 Min: 66.11 / Avg: 502.25 / Max: 723.91 Min: 21.86 / Avg: 203.92 / Max: 294.28 Min: 40.47 / Avg: 369.54 / Max: 546.36 Min: 73.14 / Avg: 287.37 / Max: 372.13 Min: 203.58 / Avg: 542.52 / Max: 747.25 Min: 76.26 / Avg: 283.96 / Max: 374.08 Min: 193.14 / Avg: 539.15 / Max: 749.93 Min: 60.35 / Avg: 271.32 / Max: 366.61 Min: 177.66 / Avg: 501.77 / Max: 743.78 Min: 26.29 / Avg: 249.88 / Max: 378.95 Min: 72.33 / Avg: 443.88 / Max: 736.09
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.43 Operation: Sharpen EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 1.20, N = 3 SE +/- 0.00, N = 3 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 SE +/- 3.00, N = 3 SE +/- 0.00, N = 3 SE +/- 1.20, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 1.67, N = 3 SE +/- 0.00, N = 3 SE +/- 1.67, N = 3 112 105 173 262 201 297 207 310 114 199 130 227 157 265 167 284 1. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp Iterations Per Minute Per Watt
OpenBenchmarking.org Iterations Per Minute Per Watt, More Is Better GraphicsMagick 1.3.43 Operation: Sharpen EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.2034 0.4068 0.6102 0.8136 1.017 0.881 0.904 0.674 0.544 0.589 0.492 0.838 0.690 0.345 0.301 0.405 0.343 0.476 0.419 0.512 0.453 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better GraphicsMagick 1.3.43 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 17.43 / Avg: 127.19 / Max: 145.11 Min: 15.82 / Avg: 116.15 / Max: 127.18 Min: 28.8 / Avg: 256.57 / Max: 278.61 Min: 59.05 / Avg: 481.76 / Max: 531.74 Min: 34.61 / Avg: 341.04 / Max: 370.17 Min: 65.27 / Avg: 603.16 / Max: 655.4 Min: 22 / Avg: 247.13 / Max: 273.84 Min: 38.07 / Avg: 448.99 / Max: 494.64 Min: 74.41 / Avg: 330.11 / Max: 353.01 Min: 197.85 / Avg: 660.39 / Max: 706.79 Min: 74.28 / Avg: 321.22 / Max: 366.4 Min: 206.23 / Avg: 662.12 / Max: 708.79 Min: 62.29 / Avg: 329.54 / Max: 363.35 Min: 148.52 / Avg: 632.93 / Max: 684.15 Min: 26.5 / Avg: 326.08 / Max: 362.61 Min: 74.46 / Avg: 627.19 / Max: 683.79
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Face Detection FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 SE +/- 0.18, N = 3 SE +/- 0.12, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.43, N = 3 SE +/- 0.43, N = 3 SE +/- 0.18, N = 3 SE +/- 0.03, N = 3 SE +/- 0.15, N = 3 SE +/- 0.40, N = 3 SE +/- 0.47, N = 3 SE +/- 0.35, N = 3 SE +/- 0.28, N = 3 SE +/- 0.00, N = 3 SE +/- 0.69, N = 3 SE +/- 0.33, N = 3 592.22 654.83 510.10 253.63 471.98 233.63 551.07 275.53 236.52 256.63 257.50 278.52 356.38 283.11 224.64 234.24 MIN: 486.17 / MAX: 612.41 MIN: 592.1 / MAX: 665.73 MIN: 263.11 / MAX: 527.97 MIN: 208.02 / MAX: 272.56 MIN: 439.42 / MAX: 492.74 MIN: 208.22 / MAX: 306.45 MIN: 513.17 / MAX: 582.25 MIN: 260.9 / MAX: 318.85 MIN: 172.47 / MAX: 265.77 MIN: 191.95 / MAX: 301.21 MIN: 187.62 / MAX: 295.06 MIN: 196.58 / MAX: 401.59 MIN: 109.11 / MAX: 395.69 MIN: 203.37 / MAX: 332.08 MIN: 153.29 / MAX: 553.64 MIN: 148.68 / MAX: 448.54 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.22 / Avg: 153 / Max: 168.49 Min: 17.81 / Avg: 126.81 / Max: 139.78 Min: 28.86 / Avg: 317.58 / Max: 349.47 Min: 53.33 / Avg: 629.45 / Max: 695.56 Min: 35.31 / Avg: 374.05 / Max: 411.15 Min: 67.22 / Avg: 740.61 / Max: 819.79 Min: 20.35 / Avg: 295.9 / Max: 332.78 Min: 42.88 / Avg: 582.16 / Max: 663.93 Min: 73.05 / Avg: 326.77 / Max: 376.97 Min: 204.33 / Avg: 649.97 / Max: 743.7 Min: 82.85 / Avg: 327.08 / Max: 376.47 Min: 200.71 / Avg: 649.1 / Max: 748.61 Min: 58.75 / Avg: 323.68 / Max: 377.26 Min: 166.52 / Avg: 639.29 / Max: 750.58 Min: 35.2 / Avg: 319.39 / Max: 384.58 Min: 74.99 / Avg: 625.11 / Max: 757.77
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.4 Harness: Deconvolution Batch shapes_3d - Engine: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.4044 0.8088 1.2132 1.6176 2.022 SE +/- 0.000791, N = 9 SE +/- 0.001685, N = 9 SE +/- 0.001484, N = 9 SE +/- 0.001921, N = 9 SE +/- 0.001272, N = 9 SE +/- 0.002070, N = 9 SE +/- 0.001089, N = 9 SE +/- 0.003127, N = 9 SE +/- 0.000508, N = 9 SE +/- 0.000806, N = 9 SE +/- 0.000927, N = 9 SE +/- 0.001240, N = 9 SE +/- 0.000805, N = 9 SE +/- 0.001274, N = 9 SE +/- 0.000392, N = 9 SE +/- 0.000902, N = 9 1.498350 1.797490 1.154870 1.009473 1.119230 1.001914 0.943621 0.829653 1.264070 0.945075 1.196430 0.864358 1.077430 0.736170 0.907161 0.624250 MIN: 1.37 MIN: 1.65 MIN: 1.04 MIN: 0.94 MIN: 1.02 MIN: 0.92 MIN: 0.92 MIN: 0.65 MIN: 1.21 MIN: 0.9 MIN: 1.16 MIN: 0.81 MIN: 1.06 MIN: 0.66 MIN: 0.87 MIN: 0.58 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better oneDNN 3.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 16.93 / Avg: 70.08 / Max: 140.81 Min: 15.81 / Avg: 62.56 / Max: 121.69 Min: 26.3 / Avg: 135.34 / Max: 274.99 Min: 53.59 / Avg: 275.56 / Max: 567.73 Min: 32.75 / Avg: 155.62 / Max: 314.15 Min: 62.97 / Avg: 300.29 / Max: 621 Min: 20.19 / Avg: 122.35 / Max: 259.22 Min: 40.43 / Avg: 245.22 / Max: 491.94 Min: 73.57 / Avg: 201.22 / Max: 367.75 Min: 194.92 / Avg: 425.54 / Max: 706.02 Min: 69.85 / Avg: 202.05 / Max: 376 Min: 187.54 / Avg: 431.59 / Max: 733.22 Min: 51.81 / Avg: 187.58 / Max: 359.76 Min: 169.45 / Avg: 409.27 / Max: 716.88 Min: 23.58 / Avg: 173.49 / Max: 373.63 Min: 70.82 / Avg: 355.72 / Max: 740.88
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.17 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200 400 600 800 1000 SE +/- 4.84, N = 4 SE +/- 6.13, N = 3 SE +/- 9.50, N = 15 SE +/- 4.37, N = 15 SE +/- 6.60, N = 3 SE +/- 7.76, N = 15 SE +/- 2.19, N = 3 SE +/- 2.36, N = 11 SE +/- 12.91, N = 15 SE +/- 1.83, N = 3 SE +/- 7.34, N = 3 SE +/- 6.21, N = 5 SE +/- 11.46, N = 3 SE +/- 6.50, N = 15 SE +/- 2.43, N = 3 SE +/- 8.88, N = 3 443.06 446.69 604.05 444.85 672.92 446.81 382.10 316.11 728.47 604.76 705.61 581.77 852.33 696.98 908.19 754.81 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.712 1.424 2.136 2.848 3.56 SE +/- 0.02450, N = 4 SE +/- 0.03117, N = 3 SE +/- 0.02493, N = 15 SE +/- 0.02209, N = 15 SE +/- 0.01470, N = 3 SE +/- 0.03659, N = 15 SE +/- 0.01501, N = 3 SE +/- 0.02347, N = 11 SE +/- 0.02719, N = 15 SE +/- 0.00498, N = 3 SE +/- 0.01456, N = 3 SE +/- 0.01853, N = 5 SE +/- 0.01594, N = 3 SE +/- 0.01405, N = 15 SE +/- 0.00291, N = 3 SE +/- 0.01541, N = 3 2.25716 2.23868 1.66046 2.25020 1.48596 2.24645 2.61653 3.16441 1.37831 1.65270 1.41662 1.71875 1.17292 1.43573 1.10042 1.32454 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 57 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2000M 4000M 6000M 8000M 10000M SE +/- 1959875.28, N = 3 SE +/- 4481443.22, N = 3 SE +/- 2618735.91, N = 3 SE +/- 60380193.59, N = 3 SE +/- 8144527.82, N = 3 SE +/- 50144890.07, N = 3 SE +/- 12739745.33, N = 3 SE +/- 8069352.17, N = 3 SE +/- 4088330.28, N = 3 SE +/- 17558125.69, N = 3 SE +/- 1976810.00, N = 3 SE +/- 28453880.81, N = 3 SE +/- 5292237.50, N = 3 SE +/- 2331189.49, N = 3 SE +/- 3555434.03, N = 3 SE +/- 61971606.40, N = 3 3116933333 2793700000 5266466667 7411866667 5811900000 8012700000 6401366667 6316266667 2848933333 4825133333 3222533333 5036000000 3500833333 5361266667 3806833333 6284300000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 57 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5M 10M 15M 20M 25M 21345877.02 23457026.73 17475938.77 12864302.70 16030850.56 11934150.92 22424548.82 16347865.87 8954185.90 7597124.43 10156047.28 7871904.97 11085612.93 8419315.63 12268330.08 10144378.39 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.88 / Avg: 146.02 / Max: 166.38 Min: 18.06 / Avg: 119.1 / Max: 135.86 Min: 30.72 / Avg: 301.36 / Max: 345.8 Min: 54.92 / Avg: 576.16 / Max: 661.98 Min: 35.02 / Avg: 362.54 / Max: 411.45 Min: 62.88 / Avg: 671.41 / Max: 791.35 Min: 22.44 / Avg: 285.46 / Max: 328.97 Min: 40.8 / Avg: 386.37 / Max: 445.76 Min: 75.93 / Avg: 318.17 / Max: 376.27 Min: 195.33 / Avg: 635.13 / Max: 752.27 Min: 75.49 / Avg: 317.3 / Max: 376.02 Min: 193.94 / Avg: 639.74 / Max: 752.86 Min: 58.98 / Avg: 315.8 / Max: 377.13 Min: 155.13 / Avg: 636.78 / Max: 752.99 Min: 24.86 / Avg: 310.3 / Max: 377.83 Min: 68.73 / Avg: 619.49 / Max: 754.72
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8 16 24 32 40 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.18, N = 15 SE +/- 0.84, N = 15 SE +/- 0.28, N = 5 SE +/- 1.06, N = 15 SE +/- 0.07, N = 3 SE +/- 1.12, N = 15 SE +/- 0.10, N = 3 SE +/- 0.16, N = 15 SE +/- 0.04, N = 3 SE +/- 0.28, N = 15 SE +/- 0.07, N = 3 SE +/- 0.12, N = 3 SE +/- 0.05, N = 3 SE +/- 0.30, N = 15 13.81 13.00 21.32 29.61 24.69 33.08 23.76 32.26 11.86 16.91 12.50 16.64 13.46 18.70 14.74 19.91 MIN: 11.85 / MAX: 16.24 MIN: 11.12 / MAX: 14.81 MIN: 18.08 / MAX: 26.9 MIN: 22.93 / MAX: 46.21 MIN: 20.63 / MAX: 30.25 MIN: 24.95 / MAX: 50.9 MIN: 20.86 / MAX: 33.07 MIN: 23.52 / MAX: 56.17 MIN: 9.93 / MAX: 13.28 MIN: 12.9 / MAX: 23.07 MIN: 10.38 / MAX: 14.93 MIN: 12.39 / MAX: 24.93 MIN: 11.09 / MAX: 16.56 MIN: 16.22 / MAX: 25.05 MIN: 12.17 / MAX: 18.21 MIN: 15 / MAX: 30.59 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0227 0.0454 0.0681 0.0908 0.1135 0.092 0.101 0.072 0.051 0.066 0.046 0.080 0.059 0.036 0.026 0.038 0.025 0.042 0.029 0.046 0.031 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 150 300 450 600 750 Min: 19.16 / Avg: 150.1 / Max: 167.85 Min: 17.62 / Avg: 128.09 / Max: 142.09 Min: 27.54 / Avg: 296.6 / Max: 330.72 Min: 57.15 / Avg: 586.21 / Max: 670.16 Min: 33.09 / Avg: 373.72 / Max: 411.13 Min: 66.69 / Avg: 722.97 / Max: 825.83 Min: 22.04 / Avg: 298.61 / Max: 336.88 Min: 41.55 / Avg: 551.09 / Max: 677.8 Min: 75.77 / Avg: 326.79 / Max: 375.2 Min: 204.79 / Avg: 656.04 / Max: 751.44 Min: 83.75 / Avg: 327.46 / Max: 377.5 Min: 201.92 / Avg: 656.25 / Max: 748.81 Min: 61.77 / Avg: 323.08 / Max: 377.72 Min: 172.79 / Avg: 646.27 / Max: 748.56 Min: 34.75 / Avg: 320.73 / Max: 377.48 Min: 82.44 / Avg: 639.12 / Max: 751.58
GPAW GPAW is a density-functional theory (DFT) Python code based on the projector-augmented wave (PAW) method and the atomic simulation environment (ASE). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better GPAW 23.6 Input: Carbon Nanotube EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 15 30 45 60 75 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.21, N = 3 SE +/- 0.29, N = 15 SE +/- 0.11, N = 3 SE +/- 0.30, N = 15 SE +/- 0.14, N = 3 SE +/- 0.18, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.27, N = 3 SE +/- 0.24, N = 3 SE +/- 0.55, N = 3 SE +/- 0.12, N = 3 SE +/- 0.10, N = 3 63.80 66.71 34.87 25.84 30.00 24.13 33.12 27.15 54.38 42.87 48.49 43.91 47.78 41.56 43.05 38.05 1. (CC) gcc options: -shared -lxc -lblas -lmpi CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better GPAW 23.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.71 / Avg: 154.54 / Max: 175.1 Min: 18.16 / Avg: 136.93 / Max: 153.06 Min: 28.47 / Avg: 300.58 / Max: 350.83 Min: 57.4 / Avg: 562.66 / Max: 711.92 Min: 34.27 / Avg: 353.01 / Max: 409.6 Min: 65.21 / Avg: 641.29 / Max: 819.6 Min: 23.61 / Avg: 293.6 / Max: 347.81 Min: 43.78 / Avg: 513.29 / Max: 700.14 Min: 72.36 / Avg: 327.19 / Max: 373.7 Min: 203.73 / Avg: 621.08 / Max: 746.35 Min: 75.93 / Avg: 324.94 / Max: 375.46 Min: 213.42 / Avg: 622.05 / Max: 746.91 Min: 59.13 / Avg: 322.66 / Max: 373.01 Min: 153.52 / Avg: 606.84 / Max: 749.04 Min: 24.08 / Avg: 314.73 / Max: 374.48 Min: 80.11 / Avg: 568.78 / Max: 750.58
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Noise Suppression Poconet-Like FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2K 4K 6K 8K 10K SE +/- 0.64, N = 3 SE +/- 3.67, N = 3 SE +/- 67.22, N = 14 SE +/- 77.59, N = 3 SE +/- 1.84, N = 3 SE +/- 59.51, N = 3 SE +/- 139.41, N = 12 SE +/- 27.90, N = 3 SE +/- 15.13, N = 3 SE +/- 37.79, N = 3 SE +/- 7.60, N = 3 SE +/- 28.98, N = 3 SE +/- 1.87, N = 3 SE +/- 13.31, N = 3 SE +/- 10.14, N = 3 SE +/- 24.76, N = 3 3636.33 3405.94 5697.51 7810.65 6963.31 8231.72 5310.07 6513.36 4208.49 5565.23 4386.48 5933.65 4209.10 7356.63 6757.26 9360.89 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.17 Model: super-resolution-10 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 60 120 180 240 300 SE +/- 0.30, N = 3 SE +/- 0.30, N = 3 SE +/- 0.67, N = 3 SE +/- 0.61, N = 3 SE +/- 0.58, N = 3 SE +/- 2.15, N = 12 SE +/- 0.26, N = 3 SE +/- 2.46, N = 15 SE +/- 0.09, N = 3 SE +/- 2.30, N = 4 SE +/- 0.25, N = 3 SE +/- 1.02, N = 3 SE +/- 0.16, N = 3 SE +/- 4.15, N = 15 SE +/- 0.03, N = 3 SE +/- 3.39, N = 3 106.53 102.97 122.68 95.84 122.95 119.56 106.13 98.64 217.88 205.89 213.99 199.90 227.42 212.92 259.19 251.31 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: super-resolution-10 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 SE +/- 0.02616, N = 3 SE +/- 0.02847, N = 3 SE +/- 0.04482, N = 3 SE +/- 0.06617, N = 3 SE +/- 0.03798, N = 3 SE +/- 0.18371, N = 12 SE +/- 0.02331, N = 3 SE +/- 0.26893, N = 15 SE +/- 0.00187, N = 3 SE +/- 0.05347, N = 4 SE +/- 0.00532, N = 3 SE +/- 0.02558, N = 3 SE +/- 0.00313, N = 3 SE +/- 0.10855, N = 15 SE +/- 0.00044, N = 3 SE +/- 0.05342, N = 3 9.38681 9.71131 8.15092 10.43440 8.13326 8.39969 9.42177 10.23106 4.58912 4.85818 4.67269 5.00222 4.39649 4.72530 3.85774 3.98013 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Memcached Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:100 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2M 4M 6M 8M 10M SE +/- 36366.89, N = 3 SE +/- 41123.40, N = 3 SE +/- 29027.89, N = 3 SE +/- 83562.82, N = 3 SE +/- 58081.25, N = 3 SE +/- 27143.58, N = 3 SE +/- 38372.96, N = 3 SE +/- 14591.70, N = 3 SE +/- 21428.06, N = 3 SE +/- 26891.33, N = 3 SE +/- 52575.97, N = 3 SE +/- 24083.78, N = 3 SE +/- 25335.31, N = 3 SE +/- 50660.01, N = 15 SE +/- 35182.67, N = 3 SE +/- 35300.21, N = 3 6198256.72 5820618.18 8220895.45 8498306.78 8634258.39 8683410.79 7736018.74 7645325.70 4384040.52 3373323.92 4315467.17 3392842.38 4481667.95 3245908.40 4752673.79 3780240.56 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre Ops/sec Per Watt
OpenBenchmarking.org Ops/sec Per Watt, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:100 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 11K 22K 33K 44K 55K 44935.74 51146.06 31074.26 17990.14 25824.73 15985.01 32910.93 19415.88 15062.13 6127.33 14272.39 5975.55 15564.53 5947.33 18089.30 7668.81 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Memcached 1.6.19 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 Min: 19.03 / Avg: 137.94 / Max: 159.73 Min: 16.61 / Avg: 113.8 / Max: 137.75 Min: 26.63 / Avg: 264.56 / Max: 307.16 Min: 55.17 / Avg: 472.39 / Max: 548.36 Min: 33.48 / Avg: 334.34 / Max: 390.93 Min: 64.05 / Avg: 543.22 / Max: 637.08 Min: 21.61 / Avg: 235.06 / Max: 274.9 Min: 40.14 / Avg: 393.77 / Max: 460.35 Min: 70.74 / Avg: 291.06 / Max: 331.31 Min: 197.32 / Avg: 550.54 / Max: 613.93 Min: 56.01 / Avg: 302.36 / Max: 343.09 Min: 189.98 / Avg: 567.79 / Max: 640.73 Min: 53.66 / Avg: 287.94 / Max: 334.72 Min: 165.51 / Avg: 545.78 / Max: 623.84 Min: 24.01 / Avg: 262.73 / Max: 308.93 Min: 62.89 / Avg: 492.94 / Max: 577.85
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.1643 0.3286 0.4929 0.6572 0.8215 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 7 SE +/- 0.00, N = 3 SE +/- 0.00, N = 9 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 0.53 0.60 0.48 0.35 0.45 0.35 0.73 0.41 0.33 0.28 0.37 0.30 0.43 0.29 0.49 0.46 MIN: 0.26 / MAX: 22.12 MIN: 0.26 / MAX: 24.7 MIN: 0.22 / MAX: 18.92 MIN: 0.32 / MAX: 81.66 MIN: 0.22 / MAX: 19.11 MIN: 0.32 / MAX: 82.66 MIN: 0.26 / MAX: 23.1 MIN: 0.39 / MAX: 91.48 MIN: 0.26 / MAX: 19.2 MIN: 0.25 / MAX: 41.4 MIN: 0.28 / MAX: 40.85 MIN: 0.26 / MAX: 42.1 MIN: 0.23 / MAX: 17.03 MIN: 0.24 / MAX: 39.69 MIN: 0.18 / MAX: 16.07 MIN: 0.18 / MAX: 22.29 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 19.35 / Avg: 152.78 / Max: 164.84 Min: 16.42 / Avg: 129.12 / Max: 139.81 Min: 26.97 / Avg: 310.58 / Max: 341.7 Min: 57.78 / Avg: 545.61 / Max: 614.54 Min: 34.06 / Avg: 367.01 / Max: 400.16 Min: 36.23 / Avg: 623.56 / Max: 689.59 Min: 21.53 / Avg: 310.47 / Max: 338.08 Min: 40.6 / Avg: 436.99 / Max: 487.81 Min: 73.52 / Avg: 330.72 / Max: 377.05 Min: 201.96 / Avg: 633.47 / Max: 674.77 Min: 83.36 / Avg: 331.78 / Max: 377.93 Min: 200.92 / Avg: 627.67 / Max: 669.86 Min: 59.31 / Avg: 328.19 / Max: 373.29 Min: 169.81 / Avg: 585.54 / Max: 631.87 Min: 33.52 / Avg: 235.56 / Max: 293.33 Min: 81.83 / Avg: 472.9 / Max: 559.21
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Person Re-Identification Retail FP16 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 8.59 9.46 7.23 3.98 6.93 3.78 8.03 4.37 7.13 7.32 7.76 7.97 9.79 7.72 6.72 6.95 MIN: 5.08 / MAX: 17.25 MIN: 5.55 / MAX: 22.41 MIN: 3.57 / MAX: 20.82 MIN: 3.41 / MAX: 18.88 MIN: 3.55 / MAX: 50.69 MIN: 3.43 / MAX: 19.66 MIN: 4.21 / MAX: 29.66 MIN: 4.13 / MAX: 27.79 MIN: 6.64 / MAX: 12.22 MIN: 6.78 / MAX: 18.72 MIN: 6.86 / MAX: 12.92 MIN: 7.04 / MAX: 21.51 MIN: 4.37 / MAX: 18.18 MIN: 7.16 / MAX: 18.31 MIN: 6.22 / MAX: 11.15 MIN: 6.28 / MAX: 20.42 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.98 / Avg: 159.92 / Max: 172.73 Min: 18.47 / Avg: 133.6 / Max: 149.13 Min: 29.57 / Avg: 331.39 / Max: 358.09 Min: 60.33 / Avg: 641.88 / Max: 698.45 Min: 33.67 / Avg: 380.02 / Max: 410.65 Min: 64.76 / Avg: 750.99 / Max: 819.5 Min: 23.26 / Avg: 322.4 / Max: 353.88 Min: 41.99 / Avg: 631.93 / Max: 700.82 Min: 73.88 / Avg: 330.38 / Max: 377.07 Min: 198.29 / Avg: 659.86 / Max: 754.53 Min: 83.77 / Avg: 330.51 / Max: 377.23 Min: 209.53 / Avg: 660.24 / Max: 754.29 Min: 57.71 / Avg: 328.31 / Max: 374.69 Min: 172.39 / Avg: 651.14 / Max: 755.22 Min: 34.07 / Avg: 323.98 / Max: 378.14 Min: 84.79 / Avg: 638.38 / Max: 770.92
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 512 - Model: AlexNet EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 400 800 1200 1600 2000 SE +/- 0.35, N = 3 SE +/- 0.57, N = 3 SE +/- 0.67, N = 3 SE +/- 12.71, N = 3 SE +/- 0.49, N = 3 SE +/- 23.15, N = 4 SE +/- 4.07, N = 3 SE +/- 7.97, N = 3 SE +/- 3.89, N = 3 SE +/- 17.88, N = 4 SE +/- 5.13, N = 3 SE +/- 8.55, N = 3 SE +/- 0.51, N = 3 SE +/- 18.52, N = 3 SE +/- 1.84, N = 3 SE +/- 19.23, N = 3 875.23 796.82 1363.33 1862.79 1392.91 1897.64 1620.53 1876.66 1047.96 1604.11 1108.95 1677.62 1278.45 1899.64 1486.12 2039.63 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 512 - Model: AlexNet EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 6.494 6.990 5.294 3.853 4.675 3.526 6.899 4.569 3.220 2.537 3.409 2.674 4.426 3.147 4.861 3.903 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.16.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 19.53 / Avg: 134.77 / Max: 159.36 Min: 17.15 / Avg: 114 / Max: 136.39 Min: 29.56 / Avg: 257.54 / Max: 304.86 Min: 59.19 / Avg: 483.44 / Max: 576.3 Min: 32.89 / Avg: 297.93 / Max: 351.95 Min: 64.96 / Avg: 538.13 / Max: 644.98 Min: 21.12 / Avg: 234.91 / Max: 276.8 Min: 42.21 / Avg: 410.73 / Max: 503.82 Min: 73.05 / Avg: 325.43 / Max: 363.64 Min: 200.65 / Avg: 632.22 / Max: 705.95 Min: 83.71 / Avg: 325.32 / Max: 365.52 Min: 209.29 / Avg: 625.4 / Max: 706.28 Min: 60.6 / Avg: 288.85 / Max: 359.42 Min: 161.55 / Avg: 603.65 / Max: 700.57 Min: 33.58 / Avg: 305.73 / Max: 353.3 Min: 84.88 / Avg: 522.57 / Max: 631.68
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.43 Operation: Enhanced EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90 180 270 360 450 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 1.20, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 2.60, N = 3 SE +/- 0.00, N = 3 SE +/- 1.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 1.15, N = 3 172 163 264 368 300 408 304 403 160 269 181 301 216 342 228 374 1. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp Iterations Per Minute Per Watt
OpenBenchmarking.org Iterations Per Minute Per Watt, More Is Better GraphicsMagick 1.3.43 Operation: Enhanced EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.3308 0.6616 0.9924 1.3232 1.654 1.348 1.470 1.067 0.815 0.935 0.742 1.342 1.014 0.484 0.410 0.549 0.466 0.656 0.568 0.696 0.614 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better GraphicsMagick 1.3.43 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 18.4 / Avg: 127.55 / Max: 140 Min: 16.93 / Avg: 110.88 / Max: 122.16 Min: 26.64 / Avg: 247.53 / Max: 267.87 Min: 59.43 / Avg: 451.41 / Max: 489.55 Min: 33.82 / Avg: 320.71 / Max: 343.53 Min: 64.84 / Avg: 549.68 / Max: 597.34 Min: 20.16 / Avg: 226.47 / Max: 243.63 Min: 40.42 / Avg: 397.41 / Max: 433.85 Min: 73.74 / Avg: 330.74 / Max: 355.56 Min: 198.68 / Avg: 656.05 / Max: 702.16 Min: 75.94 / Avg: 329.68 / Max: 362.6 Min: 202.79 / Avg: 645.53 / Max: 683.28 Min: 61.03 / Avg: 329.43 / Max: 359.83 Min: 161.82 / Avg: 601.76 / Max: 650.47 Min: 28.79 / Avg: 327.64 / Max: 359.97 Min: 74.15 / Avg: 608.66 / Max: 666.76
LAMMPS Molecular Dynamics Simulator LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: Rhodopsin Protein EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 12 24 36 48 60 SE +/- 0.29, N = 15 SE +/- 0.30, N = 15 SE +/- 0.14, N = 11 SE +/- 0.39, N = 7 SE +/- 0.35, N = 15 SE +/- 0.43, N = 7 SE +/- 0.29, N = 10 SE +/- 0.24, N = 5 SE +/- 0.05, N = 11 SE +/- 0.11, N = 9 SE +/- 0.04, N = 11 SE +/- 0.09, N = 8 SE +/- 0.36, N = 15 SE +/- 0.33, N = 8 SE +/- 0.29, N = 15 SE +/- 0.97, N = 15 24.25 21.94 36.15 50.12 40.00 53.77 40.39 51.63 27.63 47.51 30.84 52.24 28.72 42.28 34.30 55.50 1. (CXX) g++ options: -O3 -lm -ldl ns/day Per Watt
OpenBenchmarking.org ns/day Per Watt, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: Rhodopsin Protein EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.104 0.208 0.312 0.416 0.52 0.462 0.451 0.372 0.189 0.358 0.179 0.451 0.204 0.184 0.129 0.209 0.136 0.209 0.119 0.341 0.196 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 16.36 / Avg: 52.44 / Max: 142.05 Min: 15.52 / Avg: 48.67 / Max: 132.86 Min: 26.87 / Avg: 97.2 / Max: 234.7 Min: 56.67 / Avg: 265.38 / Max: 605.84 Min: 27.48 / Avg: 111.71 / Max: 349.02 Min: 60.69 / Avg: 300.13 / Max: 690.91 Min: 20.56 / Avg: 89.57 / Max: 288.34 Min: 41.5 / Avg: 252.99 / Max: 566.7 Min: 74.62 / Avg: 150.09 / Max: 352.27 Min: 188.46 / Avg: 368.48 / Max: 640.19 Min: 73.75 / Avg: 147.45 / Max: 350.72 Min: 185.06 / Avg: 384.54 / Max: 575.71 Min: 59.51 / Avg: 137.71 / Max: 356.46 Min: 158.94 / Avg: 354.59 / Max: 652.14 Min: 22.32 / Avg: 100.63 / Max: 327.13 Min: 63.52 / Avg: 283.35 / Max: 590.85
CloverLeaf CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf 1.3 Input: clover_bm64_short EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 13 26 39 52 65 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 2.15, N = 15 SE +/- 1.11, N = 12 SE +/- 1.81, N = 15 SE +/- 0.23, N = 3 SE +/- 2.07, N = 12 SE +/- 0.30, N = 3 SE +/- 0.15, N = 3 SE +/- 0.31, N = 15 SE +/- 0.17, N = 3 SE +/- 0.40, N = 4 SE +/- 0.02, N = 3 SE +/- 0.24, N = 3 SE +/- 0.03, N = 3 SE +/- 0.45, N = 3 57.38 57.54 46.50 30.05 42.17 22.83 48.36 27.74 41.43 32.25 42.22 32.44 39.48 26.58 39.38 31.90 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better CloverLeaf 1.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 19.22 / Avg: 132.55 / Max: 146.42 Min: 17.1 / Avg: 122.03 / Max: 133.14 Min: 25.91 / Avg: 223.18 / Max: 288.87 Min: 56.35 / Avg: 437.97 / Max: 551.54 Min: 33.05 / Avg: 264.63 / Max: 353.22 Min: 68.07 / Avg: 499.33 / Max: 638.11 Min: 19.77 / Avg: 183.89 / Max: 245.13 Min: 42.14 / Avg: 366.92 / Max: 461.67 Min: 76.4 / Avg: 324.2 / Max: 367.03 Min: 195.63 / Avg: 623.99 / Max: 717.07 Min: 74.72 / Avg: 323.64 / Max: 372.82 Min: 199.64 / Avg: 627.92 / Max: 713.69 Min: 57.26 / Avg: 320.54 / Max: 366.56 Min: 166.73 / Avg: 611.92 / Max: 712.51 Min: 32.08 / Avg: 299.62 / Max: 336.85 Min: 75.65 / Avg: 552.24 / Max: 662.73
Timed Node.js Compilation This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 21.7.2 Time To Compile EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 60 120 180 240 300 SE +/- 0.17, N = 3 SE +/- 0.19, N = 3 SE +/- 0.59, N = 3 SE +/- 0.18, N = 3 SE +/- 0.50, N = 3 SE +/- 1.19, N = 3 SE +/- 0.72, N = 3 SE +/- 1.51, N = 3 SE +/- 0.53, N = 3 SE +/- 0.84, N = 3 SE +/- 0.74, N = 3 SE +/- 0.52, N = 3 SE +/- 0.71, N = 3 SE +/- 0.48, N = 3 SE +/- 0.52, N = 3 SE +/- 0.12, N = 3 277.53 288.12 164.02 118.90 147.84 114.91 186.10 147.74 270.38 175.63 252.65 171.17 254.00 172.99 214.43 145.41 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed Node.js Compilation 21.7.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.11 / Avg: 134.67 / Max: 172.23 Min: 16.32 / Avg: 118.09 / Max: 147.17 Min: 29.13 / Avg: 242.47 / Max: 351.76 Min: 55.68 / Avg: 430.05 / Max: 696.5 Min: 30.95 / Avg: 289.15 / Max: 411.31 Min: 61.4 / Avg: 490.25 / Max: 822.63 Min: 21.84 / Avg: 210.13 / Max: 338.39 Min: 41.45 / Avg: 361.3 / Max: 676 Min: 74.34 / Avg: 318.88 / Max: 376.31 Min: 200.82 / Avg: 601.52 / Max: 753.5 Min: 84.71 / Avg: 317.34 / Max: 376.11 Min: 190.08 / Avg: 584.7 / Max: 754.37 Min: 56.59 / Avg: 305.63 / Max: 376.82 Min: 157.7 / Avg: 555.35 / Max: 756.3 Min: 33.43 / Avg: 295.74 / Max: 377.74 Min: 82.39 / Avg: 529.15 / Max: 756.15
7-Zip Compression This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 170K 340K 510K 680K 850K SE +/- 268.94, N = 3 SE +/- 206.69, N = 3 SE +/- 719.89, N = 3 SE +/- 6869.20, N = 3 SE +/- 1520.59, N = 3 SE +/- 3075.95, N = 3 SE +/- 1188.96, N = 3 SE +/- 2500.52, N = 3 SE +/- 1836.55, N = 3 SE +/- 5127.15, N = 3 SE +/- 2258.81, N = 3 SE +/- 3924.87, N = 3 SE +/- 1535.53, N = 3 SE +/- 4200.62, N = 3 SE +/- 1982.97, N = 3 SE +/- 709.87, N = 3 331357 319094 549500 761734 579112 741366 575440 776968 314716 491729 342363 525514 388190 592729 441636 674239 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
Parallel BZIP2 Compression This test measures the time needed to compress a file (FreeBSD-13.0-RELEASE-amd64-memstick.img) using Parallel BZIP2 compression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Parallel BZIP2 Compression 1.1.13 FreeBSD-13.0-RELEASE-amd64-memstick.img Compression EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.5922 1.1844 1.7766 2.3688 2.961 SE +/- 0.023387, N = 15 SE +/- 0.016947, N = 15 SE +/- 0.014512, N = 15 SE +/- 0.019468, N = 15 SE +/- 0.015384, N = 15 SE +/- 0.015217, N = 15 SE +/- 0.010826, N = 15 SE +/- 0.012053, N = 15 SE +/- 0.023295, N = 15 SE +/- 0.030907, N = 15 SE +/- 0.020941, N = 15 SE +/- 0.028863, N = 15 SE +/- 0.007749, N = 10 SE +/- 0.017240, N = 15 SE +/- 0.023842, N = 15 SE +/- 0.024402, N = 15 2.200497 2.321195 1.422954 1.126329 1.369063 1.109446 1.355737 1.068398 2.632077 1.982932 2.287766 1.711296 2.144857 1.469210 1.847537 1.432362 1. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Parallel BZIP2 Compression 1.1.13 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 15.97 / Avg: 55.16 / Max: 153.03 Min: 16.36 / Avg: 50.15 / Max: 132.7 Min: 26.03 / Avg: 87.61 / Max: 297.16 Min: 52.77 / Avg: 151.26 / Max: 452.63 Min: 32.91 / Avg: 100.67 / Max: 337.76 Min: 58 / Avg: 170.11 / Max: 493.47 Min: 21.06 / Avg: 71.78 / Max: 246.91 Min: 39.33 / Avg: 126.08 / Max: 368.96 Min: 72.25 / Avg: 174.54 / Max: 371.49 Min: 199.37 / Avg: 345.8 / Max: 682.95 Min: 73.85 / Avg: 174.26 / Max: 374.82 Min: 193.42 / Avg: 340.67 / Max: 703.88 Min: 59.1 / Avg: 149.7 / Max: 374.85 Min: 162.27 / Avg: 288.67 / Max: 640.02 Min: 33.01 / Avg: 124.42 / Max: 372.52 Min: 71.11 / Avg: 201.02 / Max: 594.06
LAMMPS Molecular Dynamics Simulator LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 14 28 42 56 70 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 SE +/- 0.13, N = 3 SE +/- 0.30, N = 3 SE +/- 0.12, N = 3 SE +/- 0.19, N = 3 SE +/- 0.13, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 27.53 25.33 42.51 57.10 46.89 62.00 47.99 58.84 28.28 46.71 32.16 51.82 34.89 50.52 37.46 56.40 1. (CXX) g++ options: -O3 -lm -ldl ns/day Per Watt
OpenBenchmarking.org ns/day Per Watt, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0432 0.0864 0.1296 0.1728 0.216 0.180 0.192 0.133 0.090 0.117 0.079 0.151 0.094 0.082 0.068 0.093 0.076 0.102 0.074 0.110 0.084 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 15.99 / Avg: 153.16 / Max: 156.26 Min: 15.43 / Avg: 132.09 / Max: 137.27 Min: 29.12 / Avg: 318.64 / Max: 329.98 Min: 55.66 / Avg: 633.05 / Max: 669.84 Min: 31.69 / Avg: 399.48 / Max: 409.61 Min: 64.31 / Avg: 785.79 / Max: 819.57 Min: 19.76 / Avg: 318.23 / Max: 330.98 Min: 40.67 / Avg: 624.63 / Max: 672.65 Min: 75.25 / Avg: 345.23 / Max: 374.68 Min: 186.6 / Avg: 685.33 / Max: 753.98 Min: 77.1 / Avg: 344.05 / Max: 376.79 Min: 192.84 / Avg: 682.64 / Max: 752.01 Min: 61.2 / Avg: 343.29 / Max: 376.01 Min: 165.47 / Avg: 680.62 / Max: 754.38 Min: 22.64 / Avg: 341.9 / Max: 376.37 Min: 65.48 / Avg: 673.08 / Max: 752.97
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40K 80K 120K 160K 200K SE +/- 86.99, N = 3 SE +/- 55.52, N = 3 SE +/- 205.87, N = 3 SE +/- 1922.38, N = 15 SE +/- 162.47, N = 3 SE +/- 1640.61, N = 7 SE +/- 89.35, N = 3 SE +/- 1374.23, N = 9 SE +/- 383.22, N = 3 SE +/- 18.38, N = 3 SE +/- 1761.33, N = 3 SE +/- 1422.88, N = 15 SE +/- 75.30, N = 3 SE +/- 2494.14, N = 3 SE +/- 988.12, N = 3 SE +/- 1158.60, N = 3 87511.11 78334.92 133779.62 172688.50 138605.52 178488.03 132973.43 166151.42 135075.07 174862.56 140144.01 176900.32 115106.30 189893.31 81279.79 128566.81 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
Timed LLVM Compilation This test times how long it takes to compile/build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Ninja EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 0.20, N = 3 SE +/- 0.10, N = 3 SE +/- 0.31, N = 3 SE +/- 0.08, N = 3 SE +/- 0.27, N = 3 SE +/- 0.09, N = 3 SE +/- 0.74, N = 3 SE +/- 0.45, N = 3 SE +/- 0.19, N = 3 SE +/- 0.17, N = 3 SE +/- 0.36, N = 3 SE +/- 0.79, N = 3 SE +/- 0.07, N = 3 SE +/- 0.15, N = 3 SE +/- 0.12, N = 3 SE +/- 0.27, N = 3 205.01 215.93 135.67 94.81 124.46 89.28 146.76 101.56 212.48 134.42 197.84 128.07 187.02 122.31 165.05 106.82 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed LLVM Compilation 16.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.99 / Avg: 134 / Max: 170.95 Min: 15.37 / Avg: 117.14 / Max: 147.36 Min: 28.01 / Avg: 241.32 / Max: 332.83 Min: 52.7 / Avg: 427.31 / Max: 682.72 Min: 33.31 / Avg: 289.22 / Max: 411.83 Min: 69.22 / Avg: 491.18 / Max: 825.37 Min: 21.92 / Avg: 221.05 / Max: 334.41 Min: 41.44 / Avg: 382.15 / Max: 678.1 Min: 74.97 / Avg: 314.74 / Max: 377.51 Min: 201.26 / Avg: 591.84 / Max: 754.62 Min: 83.29 / Avg: 311.03 / Max: 377.86 Min: 206.76 / Avg: 582.56 / Max: 754.19 Min: 58.01 / Avg: 298.24 / Max: 377.21 Min: 162.49 / Avg: 544.93 / Max: 754.95 Min: 33.37 / Avg: 291.41 / Max: 377.45 Min: 81.57 / Avg: 519.72 / Max: 754.2
Timed Linux Kernel Compilation This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.8 Build: allmodconfig EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90 180 270 360 450 SE +/- 0.91, N = 3 SE +/- 0.78, N = 3 SE +/- 0.49, N = 3 SE +/- 0.49, N = 3 SE +/- 0.37, N = 3 SE +/- 0.98, N = 3 SE +/- 1.05, N = 3 SE +/- 1.31, N = 3 SE +/- 0.61, N = 3 SE +/- 0.56, N = 3 SE +/- 0.67, N = 3 SE +/- 0.73, N = 3 SE +/- 0.80, N = 3 SE +/- 0.86, N = 3 SE +/- 0.72, N = 3 SE +/- 0.56, N = 3 406.41 438.17 311.83 197.63 290.34 182.00 337.69 210.20 428.39 246.94 384.06 231.03 362.45 218.57 316.90 189.74 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed Linux Kernel Compilation 6.8 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 150 300 450 600 750 Min: 16.91 / Avg: 150.79 / Max: 171.19 Min: 16.08 / Avg: 129.6 / Max: 144.82 Min: 27.23 / Avg: 282.28 / Max: 331.17 Min: 54.45 / Avg: 525.29 / Max: 678.52 Min: 34.02 / Avg: 348.42 / Max: 413.49 Min: 63.59 / Avg: 621.64 / Max: 829.42 Min: 19.75 / Avg: 262.8 / Max: 335.9 Min: 40.54 / Avg: 484.57 / Max: 689.98 Min: 70.17 / Avg: 330.93 / Max: 375.87 Min: 200.9 / Avg: 639.43 / Max: 753.88 Min: 81.1 / Avg: 330.36 / Max: 377.46 Min: 185.35 / Avg: 634.18 / Max: 755.16 Min: 56.91 / Avg: 323.36 / Max: 378.1 Min: 162.47 / Avg: 610.89 / Max: 756.51 Min: 25.82 / Avg: 317.64 / Max: 378.37 Min: 85.15 / Avg: 585.62 / Max: 755.31
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.17 Model: bertsquad-12 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 6 12 18 24 30 SE +/- 0.08, N = 3 SE +/- 0.17, N = 3 SE +/- 0.03, N = 3 SE +/- 0.10, N = 6 SE +/- 0.04, N = 3 SE +/- 0.39, N = 15 SE +/- 0.09, N = 15 SE +/- 0.27, N = 12 SE +/- 0.20, N = 3 SE +/- 0.19, N = 3 SE +/- 0.09, N = 3 SE +/- 0.35, N = 15 SE +/- 0.07, N = 3 SE +/- 0.17, N = 3 SE +/- 0.02, N = 3 SE +/- 0.22, N = 3 12.39 11.54 13.86 10.66 13.97 11.27 11.42 10.75 21.17 16.64 20.01 17.99 20.15 16.57 25.16 22.43 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: bertsquad-12 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20 40 60 80 100 SE +/- 0.51, N = 3 SE +/- 1.26, N = 3 SE +/- 0.17, N = 3 SE +/- 0.90, N = 6 SE +/- 0.18, N = 3 SE +/- 2.80, N = 15 SE +/- 0.70, N = 15 SE +/- 2.33, N = 12 SE +/- 0.44, N = 3 SE +/- 0.68, N = 3 SE +/- 0.23, N = 3 SE +/- 1.12, N = 15 SE +/- 0.17, N = 3 SE +/- 0.62, N = 3 SE +/- 0.04, N = 3 SE +/- 0.44, N = 3 80.72 86.65 72.12 93.83 71.56 90.06 87.66 93.64 47.24 60.11 49.97 55.90 49.64 60.36 39.75 44.60 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OpenRadioss OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/ and https://github.com/OpenRadioss/ModelExchange/tree/main/Examples. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: INIVOL and Fluid Structure Interaction Drop Container EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 SE +/- 0.37, N = 3 SE +/- 0.63, N = 3 SE +/- 0.57, N = 3 SE +/- 0.14, N = 3 SE +/- 0.93, N = 3 SE +/- 0.43, N = 3 SE +/- 0.46, N = 3 SE +/- 0.12, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.19, N = 3 SE +/- 0.17, N = 3 SE +/- 0.12, N = 3 SE +/- 0.30, N = 3 176.87 187.03 100.24 85.89 94.76 79.42 106.54 90.85 160.56 114.29 146.03 110.56 140.43 107.36 128.74 100.00 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenRadioss 2023.09.15 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.75 / Avg: 147.36 / Max: 159.1 Min: 18.08 / Avg: 129.32 / Max: 137.09 Min: 27.14 / Avg: 303.45 / Max: 345.2 Min: 57.91 / Avg: 595.98 / Max: 703.76 Min: 34.68 / Avg: 370.43 / Max: 409.62 Min: 65.3 / Avg: 702.37 / Max: 819.28 Min: 21.72 / Avg: 297.41 / Max: 340.37 Min: 41.48 / Avg: 554.53 / Max: 680.58 Min: 72.45 / Avg: 335.98 / Max: 368.12 Min: 204.9 / Avg: 661.11 / Max: 752.72 Min: 75.87 / Avg: 334.64 / Max: 374.39 Min: 215.12 / Avg: 658.11 / Max: 751.86 Min: 60.99 / Avg: 331.72 / Max: 375.56 Min: 175.02 / Avg: 647.82 / Max: 749.74 Min: 23.22 / Avg: 328.15 / Max: 377.07 Min: 67.07 / Avg: 633.7 / Max: 776.53
CloverLeaf CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf 1.3 Input: clover_bm16 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 SE +/- 0.79, N = 3 SE +/- 1.08, N = 3 SE +/- 3.79, N = 9 SE +/- 0.12, N = 3 SE +/- 2.24, N = 3 SE +/- 0.59, N = 3 SE +/- 4.99, N = 9 SE +/- 1.02, N = 3 SE +/- 1.43, N = 3 SE +/- 1.56, N = 3 SE +/- 2.03, N = 3 SE +/- 2.11, N = 9 SE +/- 0.13, N = 3 SE +/- 0.74, N = 3 SE +/- 0.54, N = 3 SE +/- 2.02, N = 3 495.25 498.22 334.30 266.10 245.82 223.54 400.52 349.67 351.29 263.33 357.57 269.20 328.47 214.80 410.59 378.05 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better CloverLeaf 1.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.26 / Avg: 143.82 / Max: 147.88 Min: 18.03 / Avg: 132.66 / Max: 136.36 Min: 28.5 / Avg: 250.17 / Max: 287.09 Min: 57.96 / Avg: 483.81 / Max: 500.79 Min: 34.9 / Avg: 301.76 / Max: 361.01 Min: 66.27 / Avg: 538.72 / Max: 566.53 Min: 19.15 / Avg: 200.7 / Max: 237.69 Min: 42.42 / Avg: 388.45 / Max: 400.3 Min: 71.71 / Avg: 345.92 / Max: 368.64 Min: 198.24 / Avg: 677.8 / Max: 709.86 Min: 75.8 / Avg: 346.15 / Max: 373.49 Min: 200.21 / Avg: 680.25 / Max: 711.02 Min: 57.58 / Avg: 346.05 / Max: 371.34 Min: 171.76 / Avg: 685.99 / Max: 712.37 Min: 33.19 / Avg: 289.13 / Max: 310.83 Min: 76.61 / Avg: 524.03 / Max: 570.93
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: AlexNet EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 400 800 1200 1600 2000 SE +/- 0.99, N = 3 SE +/- 0.41, N = 3 SE +/- 2.58, N = 3 SE +/- 6.01, N = 3 SE +/- 1.54, N = 3 SE +/- 15.82, N = 3 SE +/- 4.01, N = 3 SE +/- 10.23, N = 15 SE +/- 1.21, N = 3 SE +/- 1.15, N = 3 SE +/- 6.82, N = 3 SE +/- 5.81, N = 3 SE +/- 0.54, N = 3 SE +/- 5.06, N = 3 SE +/- 4.27, N = 3 SE +/- 1.86, N = 3 843.63 771.47 1278.58 1524.74 1301.26 1560.56 1454.71 1403.46 986.04 1479.53 1058.99 1528.41 1205.30 1730.25 1277.46 1681.44 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: AlexNet EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 6.684 7.144 5.419 3.535 4.808 3.248 6.976 3.829 3.205 2.515 3.444 2.593 4.146 3.122 4.749 3.635 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.16.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 17.24 / Avg: 126.21 / Max: 154.11 Min: 18.12 / Avg: 108 / Max: 132.76 Min: 26.89 / Avg: 235.94 / Max: 296.09 Min: 54.66 / Avg: 428.39 / Max: 556.88 Min: 33.53 / Avg: 270.65 / Max: 335.43 Min: 69.6 / Avg: 480.42 / Max: 618.89 Min: 22.22 / Avg: 206.77 / Max: 269.72 Min: 40.95 / Avg: 366.52 / Max: 481.87 Min: 71.72 / Avg: 307.61 / Max: 356.17 Min: 206.23 / Avg: 588.17 / Max: 702.19 Min: 83.56 / Avg: 307.45 / Max: 358.74 Min: 196.2 / Avg: 589.54 / Max: 703.42 Min: 58.81 / Avg: 290.68 / Max: 351.63 Min: 155.31 / Avg: 554.16 / Max: 691.31 Min: 33.64 / Avg: 268.97 / Max: 332.41 Min: 79.54 / Avg: 462.52 / Max: 594.2
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.7 Preset: Medium EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 SE +/- 0.07, N = 6 SE +/- 0.07, N = 6 SE +/- 0.20, N = 7 SE +/- 3.54, N = 6 SE +/- 0.63, N = 7 SE +/- 3.11, N = 7 SE +/- 0.39, N = 7 SE +/- 0.52, N = 6 SE +/- 0.03, N = 6 SE +/- 0.52, N = 6 SE +/- 0.08, N = 6 SE +/- 0.77, N = 6 SE +/- 0.16, N = 6 SE +/- 2.79, N = 6 SE +/- 1.17, N = 6 SE +/- 1.56, N = 7 294.62 273.67 449.11 491.53 500.77 508.09 527.80 462.24 247.30 441.65 284.33 483.10 333.99 484.55 327.78 550.02 1. (CXX) g++ options: -O3 -flto -pthread MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.7 Preset: Medium EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1.1363 2.2726 3.4089 4.5452 5.6815 3.926 3.989 3.553 2.099 3.488 1.966 5.050 2.377 1.167 1.094 1.373 1.215 1.923 1.414 2.012 1.983 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.7 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 16.05 / Avg: 75.04 / Max: 166.06 Min: 15.15 / Avg: 68.61 / Max: 143.04 Min: 27.98 / Avg: 126.4 / Max: 324.02 Min: 57.21 / Avg: 234.22 / Max: 515.92 Min: 32.4 / Avg: 143.57 / Max: 393.87 Min: 60.92 / Avg: 258.49 / Max: 596.25 Min: 20.85 / Avg: 104.51 / Max: 314.72 Min: 38.51 / Avg: 194.48 / Max: 419.92 Min: 73.51 / Avg: 211.87 / Max: 372.54 Min: 202.14 / Avg: 403.63 / Max: 727.09 Min: 46.45 / Avg: 207.12 / Max: 368.04 Min: 188.21 / Avg: 397.48 / Max: 724.85 Min: 60.47 / Avg: 173.64 / Max: 368.66 Min: 149.65 / Avg: 342.67 / Max: 671.28 Min: 26.56 / Avg: 162.88 / Max: 364.95 Min: 75.77 / Avg: 277.37 / Max: 691.02
Y-Cruncher Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.8.3 Pi Digits To Calculate: 500M EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1.1621 2.3242 3.4863 4.6484 5.8105 SE +/- 0.006, N = 6 SE +/- 0.008, N = 6 SE +/- 0.003, N = 6 SE +/- 0.003, N = 5 SE +/- 0.003, N = 6 SE +/- 0.007, N = 5 SE +/- 0.005, N = 6 SE +/- 0.015, N = 5 SE +/- 0.010, N = 7 SE +/- 0.064, N = 15 SE +/- 0.023, N = 15 SE +/- 0.074, N = 15 SE +/- 0.003, N = 7 SE +/- 0.002, N = 8 SE +/- 0.008, N = 7 SE +/- 0.004, N = 7 5.077 5.165 3.850 4.329 4.312 4.510 3.968 4.100 3.781 3.110 3.704 2.961 3.267 2.364 3.391 2.659 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Y-Cruncher 0.8.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 16.15 / Avg: 86.13 / Max: 163.32 Min: 16.77 / Avg: 78.49 / Max: 143.84 Min: 27.89 / Avg: 146.25 / Max: 317.02 Min: 53.19 / Avg: 293.51 / Max: 558.3 Min: 32.78 / Avg: 173.58 / Max: 352.72 Min: 66.18 / Avg: 327.67 / Max: 622.14 Min: 20.76 / Avg: 118.84 / Max: 272.84 Min: 37.77 / Avg: 238.41 / Max: 478.35 Min: 74.95 / Avg: 212.09 / Max: 370.6 Min: 197.01 / Avg: 434.63 / Max: 747.05 Min: 70.98 / Avg: 211.16 / Max: 373.06 Min: 189.46 / Avg: 425.51 / Max: 749.34 Min: 57.55 / Avg: 181.21 / Max: 363.23 Min: 164.95 / Avg: 367.73 / Max: 733.59 Min: 31.93 / Avg: 151.06 / Max: 346.25 Min: 81.75 / Avg: 285.35 / Max: 686.94
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Handwritten English Recognition FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 10 20 30 40 50 SE +/- 0.09, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 41.38 45.71 34.92 36.90 32.00 33.18 38.85 39.86 23.12 24.00 24.94 25.72 36.20 25.12 21.87 22.96 MIN: 25.29 / MAX: 59.37 MIN: 31.06 / MAX: 64.63 MIN: 22.04 / MAX: 54.74 MIN: 29.34 / MAX: 49.06 MIN: 21.72 / MAX: 52.65 MIN: 29.34 / MAX: 51.4 MIN: 23.02 / MAX: 69.97 MIN: 35.87 / MAX: 58.79 MIN: 18.74 / MAX: 31.08 MIN: 19.19 / MAX: 45.88 MIN: 19.75 / MAX: 33.18 MIN: 20.7 / MAX: 44.98 MIN: 20.73 / MAX: 45.28 MIN: 19.86 / MAX: 59.96 MIN: 20.16 / MAX: 32.36 MIN: 20.8 / MAX: 34.86 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.94 / Avg: 165.08 / Max: 177.71 Min: 16.48 / Avg: 135.3 / Max: 146.77 Min: 30.45 / Avg: 333.47 / Max: 360.29 Min: 59.67 / Avg: 652.58 / Max: 705.65 Min: 31.71 / Avg: 382.65 / Max: 411.13 Min: 67.9 / Avg: 760.8 / Max: 819.21 Min: 23.52 / Avg: 323.53 / Max: 352.29 Min: 42.75 / Avg: 632.45 / Max: 690.87 Min: 75.49 / Avg: 331.88 / Max: 373.56 Min: 208.68 / Avg: 665.43 / Max: 751.81 Min: 87.21 / Avg: 332.74 / Max: 375.71 Min: 200.26 / Avg: 664.69 / Max: 755.11 Min: 59.81 / Avg: 328.72 / Max: 377.28 Min: 175.49 / Avg: 660.59 / Max: 741.22 Min: 24.83 / Avg: 318.97 / Max: 381.52 Min: 79.4 / Avg: 653.57 / Max: 770.7
Y-Cruncher Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.8.3 Pi Digits To Calculate: 1B EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 SE +/- 0.005, N = 4 SE +/- 0.001, N = 4 SE +/- 0.008, N = 5 SE +/- 0.012, N = 4 SE +/- 0.008, N = 4 SE +/- 0.012, N = 4 SE +/- 0.007, N = 4 SE +/- 0.049, N = 4 SE +/- 0.045, N = 5 SE +/- 0.122, N = 15 SE +/- 0.026, N = 5 SE +/- 0.142, N = 12 SE +/- 0.005, N = 5 SE +/- 0.007, N = 6 SE +/- 0.030, N = 5 SE +/- 0.010, N = 6 10.205 10.427 6.743 8.196 7.471 8.992 6.928 7.915 8.300 6.334 7.959 6.266 7.047 5.008 6.701 5.110 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Y-Cruncher 0.8.3 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 16.07 / Avg: 105.12 / Max: 171.94 Min: 16.4 / Avg: 94.34 / Max: 147.45 Min: 28.49 / Avg: 178.56 / Max: 342.72 Min: 58.9 / Avg: 351.79 / Max: 591.63 Min: 32.32 / Avg: 212.59 / Max: 399.38 Min: 66.34 / Avg: 404.17 / Max: 643.66 Min: 21.57 / Avg: 148.59 / Max: 323.95 Min: 40.66 / Avg: 284.7 / Max: 529.69 Min: 76.02 / Avg: 255.33 / Max: 373.96 Min: 191.23 / Avg: 496.54 / Max: 748.22 Min: 73.1 / Avg: 254.54 / Max: 371.3 Min: 193.6 / Avg: 493.66 / Max: 748.88 Min: 30.38 / Avg: 222.94 / Max: 366.15 Min: 166.8 / Avg: 424.33 / Max: 738.75 Min: 32.02 / Avg: 198.74 / Max: 364.57 Min: 76.49 / Avg: 362.82 / Max: 745.4
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 512 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300M 600M 900M 1200M 1500M SE +/- 390953.25, N = 3 SE +/- 431058.13, N = 3 SE +/- 1047748.91, N = 3 SE +/- 2338327.99, N = 3 SE +/- 2412697.89, N = 3 SE +/- 3901424.24, N = 3 SE +/- 814452.78, N = 3 SE +/- 503322.30, N = 3 SE +/- 681281.47, N = 3 SE +/- 11560043.25, N = 5 SE +/- 1345523.44, N = 3 SE +/- 2123152.79, N = 3 SE +/- 1740596.96, N = 3 SE +/- 6842108.19, N = 3 SE +/- 2313246.88, N = 3 SE +/- 4790383.89, N = 3 809916667 731756667 1158166667 1199966667 1368733333 1394733333 1369800000 1349800000 800186667 1091460000 882200000 1206066667 918566667 1126466667 1152933333 1493833333 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 512 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1.4M 2.8M 4.2M 5.6M 7M 5867538.62 6364836.35 4367851.07 2957086.75 3778092.65 2805690.78 5183635.11 3727692.53 2528910.04 1702250.87 2800439.16 1899608.97 2929507.99 1966021.92 3740373.11 2443783.02 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.57 / Avg: 138.03 / Max: 156.97 Min: 16.09 / Avg: 114.97 / Max: 130.99 Min: 26.08 / Avg: 265.16 / Max: 303.52 Min: 53.51 / Avg: 405.79 / Max: 476.68 Min: 32.21 / Avg: 362.28 / Max: 409.42 Min: 65.97 / Avg: 497.11 / Max: 585.32 Min: 21.47 / Avg: 264.25 / Max: 302.73 Min: 40.86 / Avg: 362.1 / Max: 421.05 Min: 77.21 / Avg: 316.42 / Max: 377.29 Min: 197.48 / Avg: 641.19 / Max: 750.61 Min: 75.73 / Avg: 315.02 / Max: 376.2 Min: 206.37 / Avg: 634.9 / Max: 748.44 Min: 60 / Avg: 313.56 / Max: 376 Min: 168.69 / Avg: 572.97 / Max: 696.31 Min: 27.09 / Avg: 308.24 / Max: 376.57 Min: 70.15 / Avg: 611.28 / Max: 739.63
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better OpenVINO 2024.0 Model: Face Detection FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.12, N = 3 SE +/- 0.41, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.14, N = 3 SE +/- 0.58, N = 3 SE +/- 0.41, N = 3 SE +/- 0.54, N = 3 SE +/- 0.12, N = 3 SE +/- 0.01, N = 3 SE +/- 0.69, N = 3 SE +/- 0.77, N = 3 53.89 48.73 93.99 188.74 101.51 205.03 115.78 231.83 202.59 373.50 217.11 401.45 167.83 423.18 284.64 545.21 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 512 - Model: GoogLeNet EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 4.02, N = 10 SE +/- 5.11, N = 3 SE +/- 3.05, N = 12 SE +/- 1.60, N = 3 SE +/- 12.03, N = 12 SE +/- 5.63, N = 6 SE +/- 0.41, N = 3 SE +/- 1.14, N = 3 SE +/- 0.37, N = 3 SE +/- 2.46, N = 3 SE +/- 0.34, N = 3 SE +/- 2.62, N = 3 SE +/- 3.68, N = 3 SE +/- 1.52, N = 3 313.50 295.29 470.95 525.84 425.25 464.78 450.61 588.43 383.57 467.91 398.40 477.33 461.23 576.08 462.78 577.76 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 512 - Model: GoogLeNet EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.5085 1.017 1.5255 2.034 2.5425 2.098 2.260 1.646 0.960 1.293 0.747 1.824 1.245 1.137 0.696 1.180 0.710 1.378 0.871 1.511 0.974 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.16.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.68 / Avg: 149.41 / Max: 158.78 Min: 16.37 / Avg: 130.65 / Max: 139.18 Min: 16.07 / Avg: 286.15 / Max: 318.45 Min: 60.16 / Avg: 547.94 / Max: 597.52 Min: 33.11 / Avg: 328.89 / Max: 367.06 Min: 65.4 / Avg: 622.34 / Max: 677.73 Min: 19.9 / Avg: 247.07 / Max: 298.24 Min: 41.61 / Avg: 472.75 / Max: 541.37 Min: 73.74 / Avg: 337.25 / Max: 358.42 Min: 205.47 / Avg: 671.88 / Max: 722 Min: 83.34 / Avg: 337.62 / Max: 362.35 Min: 204.77 / Avg: 672.55 / Max: 725.44 Min: 62.15 / Avg: 334.67 / Max: 366.71 Min: 163.66 / Avg: 661.11 / Max: 719.14 Min: 35.37 / Avg: 306.26 / Max: 336.64 Min: 81.09 / Avg: 593.47 / Max: 647.85
GNU Octave Benchmark This test profile measures how long it takes to complete several reference GNU Octave files via octave-benchmark. GNU Octave is used for numerical computations and is an open-source alternative to MATLAB. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better GNU Octave Benchmark 8.4.0 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 SE +/- 0.011, N = 5 SE +/- 0.007, N = 5 SE +/- 0.009, N = 6 SE +/- 0.004, N = 6 SE +/- 0.018, N = 6 SE +/- 0.007, N = 6 SE +/- 0.014, N = 5 SE +/- 0.025, N = 5 SE +/- 0.015, N = 6 SE +/- 0.096, N = 5 SE +/- 0.024, N = 6 SE +/- 0.082, N = 5 SE +/- 0.015, N = 6 SE +/- 0.019, N = 6 SE +/- 0.020, N = 6 SE +/- 0.019, N = 6 8.929 8.948 7.528 7.645 6.857 6.955 8.801 8.848 7.622 12.617 7.748 12.578 7.532 8.036 6.458 6.856 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better GNU Octave Benchmark 8.4.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ 80 160 240 320 400 Min: 17.25 / Avg: 49.95 / Max: 77.01 Min: 15.49 / Avg: 47.63 / Max: 72.89 Min: 27.81 / Avg: 84.26 / Max: 135.67 Min: 56.62 / Avg: 189.57 / Max: 270.57 Min: 34.49 / Avg: 95.8 / Max: 154.31 Min: 61.74 / Avg: 205.11 / Max: 301.54 Min: 20.26 / Avg: 67.77 / Max: 105.08 Min: 40.57 / Avg: 161.36 / Max: 222.38 Min: 74.85 / Avg: 164.13 / Max: 237.93 Min: 202.03 / Avg: 373.8 / Max: 446.62 Min: 79.51 / Avg: 166.55 / Max: 232.48 Min: 197.53 / Avg: 371.49 / Max: 456.41 Min: 56.04 / Avg: 133.36 / Max: 217.89 Min: 167.67 / Avg: 296.93 / Max: 375.38 Min: 32.85 / Avg: 108.34 / Max: 202.86
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.43 Operation: Swirl EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 150 300 450 600 750 SE +/- 0.33, N = 3 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 1.33, N = 3 SE +/- 0.67, N = 3 SE +/- 0.67, N = 3 SE +/- 1.15, N = 3 SE +/- 1.76, N = 3 SE +/- 0.88, N = 3 SE +/- 0.58, N = 3 379 363 511 553 546 583 535 567 394 509 432 558 502 617 529 689 1. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp Iterations Per Minute Per Watt
OpenBenchmarking.org Iterations Per Minute Per Watt, More Is Better GraphicsMagick 1.3.43 Operation: Swirl EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.7398 1.4796 2.2194 2.9592 3.699 3.093 3.288 2.290 1.445 2.036 1.298 2.779 1.710 1.193 0.851 1.341 0.964 1.617 1.137 1.724 1.307 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better GraphicsMagick 1.3.43 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 Min: 18.54 / Avg: 122.54 / Max: 132.44 Min: 17.2 / Avg: 110.39 / Max: 118.71 Min: 30.38 / Avg: 223.17 / Max: 243.96 Min: 56.5 / Avg: 382.8 / Max: 426.76 Min: 34.46 / Avg: 268.23 / Max: 288.18 Min: 62.97 / Avg: 449.12 / Max: 484.89 Min: 21.51 / Avg: 192.51 / Max: 206.12 Min: 40.54 / Avg: 331.64 / Max: 362.33 Min: 72.86 / Avg: 330.14 / Max: 351.19 Min: 191.82 / Avg: 598.34 / Max: 636.3 Min: 70.83 / Avg: 322.19 / Max: 348.63 Min: 192.72 / Avg: 578.67 / Max: 624.15 Min: 59.04 / Avg: 310.53 / Max: 334.67 Min: 159.26 / Avg: 542.66 / Max: 585.44 Min: 25.3 / Avg: 306.84 / Max: 336.43 Min: 80.88 / Avg: 527.29 / Max: 574.8
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 32 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 900M 1800M 2700M 3600M 4500M SE +/- 3138116.92, N = 3 SE +/- 2003607.86, N = 3 SE +/- 2714365.08, N = 3 SE +/- 3186603.90, N = 3 SE +/- 841295.30, N = 3 SE +/- 4250882.26, N = 3 SE +/- 4056818.68, N = 3 SE +/- 384418.75, N = 3 SE +/- 400000.00, N = 3 SE +/- 4598671.31, N = 3 SE +/- 1342054.81, N = 3 SE +/- 2403700.85, N = 3 SE +/- 1858314.65, N = 3 SE +/- 26633124.74, N = 3 SE +/- 2236316.42, N = 3 SE +/- 13813198.20, N = 3 3047066667 2767933333 4032733333 4063333333 4353533333 4407900000 3791566667 3794133333 2351900000 3254033333 2739633333 3337066667 3200600000 3558600000 3148233333 3520766667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 32 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5M 10M 15M 20M 25M 21706213.46 23574098.60 14392112.20 9803609.24 11993608.96 8852490.97 15541749.24 10891038.63 8304651.14 5098423.00 8871933.76 5205818.86 10135010.79 6918246.02 10144782.58 6353287.78 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 16.53 / Avg: 140.38 / Max: 157.26 Min: 16.54 / Avg: 117.41 / Max: 138.28 Min: 28.95 / Avg: 280.2 / Max: 318.02 Min: 56.36 / Avg: 414.47 / Max: 483.6 Min: 35.34 / Avg: 362.99 / Max: 409.42 Min: 66.83 / Avg: 497.93 / Max: 572.66 Min: 21.67 / Avg: 243.96 / Max: 276.65 Min: 40.25 / Avg: 348.37 / Max: 393.82 Min: 74.4 / Avg: 283.2 / Max: 376.87 Min: 189.62 / Avg: 638.24 / Max: 714.51 Min: 80.13 / Avg: 308.8 / Max: 366.56 Min: 196.59 / Avg: 641.03 / Max: 736.11 Min: 62.65 / Avg: 315.8 / Max: 377.95 Min: 144.18 / Avg: 514.38 / Max: 660.61 Min: 27.32 / Avg: 310.33 / Max: 359.1 Min: 75.15 / Avg: 554.16 / Max: 686.41
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: GoogLeNet EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 3.87, N = 15 SE +/- 1.02, N = 3 SE +/- 2.76, N = 15 SE +/- 0.51, N = 3 SE +/- 10.93, N = 12 SE +/- 1.11, N = 3 SE +/- 1.67, N = 3 SE +/- 2.29, N = 3 SE +/- 1.91, N = 3 SE +/- 4.80, N = 3 SE +/- 0.24, N = 3 SE +/- 4.87, N = 3 SE +/- 2.30, N = 3 SE +/- 2.68, N = 3 308.46 287.42 447.61 401.84 406.58 391.40 462.79 397.44 375.73 447.50 386.00 449.91 448.46 528.93 423.59 501.39 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: GoogLeNet EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.5108 1.0216 1.5324 2.0432 2.554 2.115 2.270 1.620 0.775 1.305 0.666 1.896 0.932 1.141 0.684 1.170 0.690 1.385 0.833 1.503 0.936 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.16.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.95 / Avg: 145.84 / Max: 158.71 Min: 17.59 / Avg: 126.6 / Max: 137.81 Min: 27.68 / Avg: 276.32 / Max: 312.83 Min: 57.21 / Avg: 518.78 / Max: 576.81 Min: 30.3 / Avg: 311.48 / Max: 350.99 Min: 65.43 / Avg: 587.49 / Max: 655.55 Min: 21.26 / Avg: 244.09 / Max: 296.43 Min: 40.92 / Avg: 426.33 / Max: 487.57 Min: 73.87 / Avg: 329.22 / Max: 366.78 Min: 201.04 / Avg: 653.82 / Max: 709.07 Min: 79.85 / Avg: 329.98 / Max: 363.65 Min: 201.04 / Avg: 651.98 / Max: 714.73 Min: 58.65 / Avg: 323.79 / Max: 362.41 Min: 161.35 / Avg: 635.04 / Max: 705.2 Min: 33.46 / Avg: 281.92 / Max: 317.12 Min: 82.2 / Avg: 535.58 / Max: 613.03
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.34, N = 3 SE +/- 1.83, N = 3 SE +/- 1.42, N = 4 SE +/- 0.31, N = 3 SE +/- 1.62, N = 4 SE +/- 0.81, N = 3 SE +/- 1.17, N = 3 SE +/- 0.79, N = 3 SE +/- 0.61, N = 3 SE +/- 1.80, N = 3 SE +/- 0.01, N = 3 SE +/- 1.39, N = 3 SE +/- 0.48, N = 3 SE +/- 0.61, N = 3 101.17 95.63 140.01 161.90 121.96 137.50 135.17 174.93 117.53 145.29 125.14 138.39 148.45 174.95 140.53 169.75 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.16 0.32 0.48 0.64 0.8 0.659 0.711 0.487 0.296 0.385 0.227 0.553 0.366 0.343 0.212 0.364 0.202 0.433 0.258 0.469 0.290 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.16.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.7 / Avg: 153.51 / Max: 165.78 Min: 16.14 / Avg: 134.5 / Max: 142.97 Min: 30.39 / Avg: 287.56 / Max: 321.33 Min: 59.62 / Avg: 547.08 / Max: 589.2 Min: 35.54 / Avg: 316.96 / Max: 353.37 Min: 66.7 / Avg: 606.04 / Max: 658.36 Min: 20.9 / Avg: 244.22 / Max: 295.24 Min: 40.89 / Avg: 478.51 / Max: 540.78 Min: 74.91 / Avg: 342.39 / Max: 362.9 Min: 207.09 / Avg: 684.53 / Max: 724.92 Min: 85.58 / Avg: 343.75 / Max: 366.49 Min: 208.64 / Avg: 684.02 / Max: 721.63 Min: 61.57 / Avg: 342.65 / Max: 363.24 Min: 161.89 / Avg: 678.16 / Max: 719.66 Min: 33.86 / Avg: 299.63 / Max: 330.67 Min: 83.43 / Avg: 585.9 / Max: 658.36
Timed Linux Kernel Compilation This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.8 Build: defconfig EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 11 22 33 44 55 SE +/- 0.50, N = 4 SE +/- 0.54, N = 4 SE +/- 0.43, N = 3 SE +/- 0.33, N = 3 SE +/- 0.46, N = 3 SE +/- 0.17, N = 13 SE +/- 0.44, N = 3 SE +/- 0.28, N = 7 SE +/- 0.59, N = 3 SE +/- 0.40, N = 4 SE +/- 0.57, N = 3 SE +/- 0.31, N = 5 SE +/- 0.55, N = 3 SE +/- 0.43, N = 3 SE +/- 0.39, N = 4 SE +/- 0.23, N = 7 44.28 46.35 34.57 27.68 32.89 25.87 38.53 30.68 43.42 32.14 40.50 30.53 39.45 29.98 34.59 25.87 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed Linux Kernel Compilation 6.8 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 16.1 / Avg: 97.22 / Max: 163.42 Min: 14.06 / Avg: 87.28 / Max: 138.54 Min: 27.91 / Avg: 175.53 / Max: 322.12 Min: 52.83 / Avg: 317.6 / Max: 662.66 Min: 32.62 / Avg: 210.93 / Max: 409.9 Min: 62.86 / Avg: 375.56 / Max: 824.56 Min: 20.78 / Avg: 155.1 / Max: 328.08 Min: 38.17 / Avg: 283.46 / Max: 659.24 Min: 73.73 / Avg: 265.23 / Max: 376.52 Min: 189.61 / Avg: 492.26 / Max: 752.95 Min: 69.92 / Avg: 260.34 / Max: 376.76 Min: 176.87 / Avg: 486.28 / Max: 754.8 Min: 60.49 / Avg: 234.11 / Max: 377.45 Min: 150.36 / Avg: 425.97 / Max: 755.88 Min: 26.35 / Avg: 214.95 / Max: 377.53 Min: 73.91 / Avg: 384.63 / Max: 756.99
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 57 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1000M 2000M 3000M 4000M 5000M SE +/- 3527510.93, N = 3 SE +/- 6398003.16, N = 3 SE +/- 20120967.50, N = 3 SE +/- 4098915.85, N = 3 SE +/- 6778479.02, N = 3 SE +/- 11892621.43, N = 3 SE +/- 18691531.06, N = 3 SE +/- 14606429.33, N = 3 SE +/- 2030052.00, N = 3 SE +/- 25888414.40, N = 3 SE +/- 3119294.79, N = 3 SE +/- 14772609.79, N = 3 SE +/- 3116800.35, N = 3 SE +/- 13655808.69, N = 3 SE +/- 5260228.13, N = 3 SE +/- 51282832.98, N = 4 2982800000 2687066667 4256300000 4280933333 4755333333 4759366667 4714500000 4626566667 2670033333 3596400000 2972400000 3896500000 3181266667 3723666667 3503200000 4450525000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 57 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5M 10M 15M 20M 25M 20620501.38 22733164.52 14722432.57 10089896.41 13085490.53 9563226.72 16705633.01 12234041.51 8437038.99 5639264.93 9312055.82 6058163.97 10089438.81 6447021.00 11273487.11 7209189.04 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.85 / Avg: 144.65 / Max: 162.72 Min: 17.47 / Avg: 118.2 / Max: 136.58 Min: 29.04 / Avg: 289.1 / Max: 326.36 Min: 57.66 / Avg: 424.28 / Max: 495.38 Min: 35.08 / Avg: 363.41 / Max: 409.45 Min: 60.6 / Avg: 497.67 / Max: 581.47 Min: 21.64 / Avg: 282.21 / Max: 320.93 Min: 40.47 / Avg: 378.17 / Max: 438.15 Min: 75.83 / Avg: 316.47 / Max: 376.18 Min: 204.5 / Avg: 637.74 / Max: 753.77 Min: 79.18 / Avg: 319.2 / Max: 376.38 Min: 195.26 / Avg: 643.18 / Max: 753.46 Min: 32.16 / Avg: 315.31 / Max: 376.7 Min: 158.24 / Avg: 577.58 / Max: 700.7 Min: 24.5 / Avg: 310.75 / Max: 376.14 Min: 69.61 / Avg: 617.34 / Max: 730.41
ClickHouse ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 SE +/- 1.77, N = 3 SE +/- 5.23, N = 3 SE +/- 4.03, N = 3 SE +/- 10.00, N = 3 SE +/- 4.34, N = 3 SE +/- 2.03, N = 3 SE +/- 3.32, N = 3 SE +/- 6.21, N = 5 SE +/- 6.03, N = 3 SE +/- 3.08, N = 3 SE +/- 3.60, N = 3 SE +/- 3.54, N = 12 SE +/- 3.45, N = 3 SE +/- 2.55, N = 3 SE +/- 3.62, N = 6 SE +/- 5.39, N = 5 338.84 335.46 559.55 525.23 595.19 526.36 458.96 427.95 470.02 501.07 476.91 471.79 481.66 501.63 458.23 438.60 MIN: 35.76 / MAX: 4000 MIN: 32.86 / MAX: 5454.55 MIN: 57.03 / MAX: 5000 MIN: 81.86 / MAX: 5000 MIN: 62.05 / MAX: 5000 MIN: 91.05 / MAX: 6000 MIN: 68.65 / MAX: 4000 MIN: 80.54 / MAX: 5454.55 MIN: 32.59 / MAX: 5454.55 MIN: 54.25 / MAX: 5000 MIN: 36.01 / MAX: 5000 MIN: 53.43 / MAX: 6000 MIN: 40.16 / MAX: 5454.55 MIN: 64.45 / MAX: 6666.67 MIN: 42.16 / MAX: 6000 MIN: 58.2 / MAX: 6000 Queries Per Minute, Geo Mean Per Watt
OpenBenchmarking.org Queries Per Minute, Geo Mean Per Watt, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.9824 1.9648 2.9472 3.9296 4.912 4.085 4.366 4.053 1.870 3.709 1.676 4.182 1.857 1.967 1.068 1.974 1.012 2.341 1.265 2.599 1.289 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better ClickHouse 22.12.3.5 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.06 / Avg: 82.96 / Max: 166.09 Min: 15.83 / Avg: 76.83 / Max: 137.9 Min: 28.3 / Avg: 138.04 / Max: 305.03 Min: 54.72 / Avg: 280.91 / Max: 531.55 Min: 29.83 / Avg: 160.48 / Max: 343.21 Min: 64.74 / Avg: 314 / Max: 643.44 Min: 20.77 / Avg: 109.76 / Max: 265.96 Min: 40.81 / Avg: 230.4 / Max: 456.57 Min: 70.28 / Avg: 238.9 / Max: 367.38 Min: 199.78 / Avg: 469.37 / Max: 713.86 Min: 77.89 / Avg: 241.61 / Max: 372.85 Min: 199.33 / Avg: 466.07 / Max: 698.95 Min: 52.84 / Avg: 205.71 / Max: 375.59 Min: 157.92 / Avg: 396.39 / Max: 673.71 Min: 26.54 / Avg: 176.29 / Max: 377.99 Min: 77.74 / Avg: 340.3 / Max: 642.4
OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 SE +/- 2.99, N = 3 SE +/- 5.72, N = 3 SE +/- 3.14, N = 3 SE +/- 8.37, N = 3 SE +/- 0.42, N = 3 SE +/- 6.78, N = 3 SE +/- 9.26, N = 3 SE +/- 4.85, N = 5 SE +/- 1.94, N = 3 SE +/- 0.19, N = 3 SE +/- 4.95, N = 3 SE +/- 3.17, N = 12 SE +/- 2.93, N = 3 SE +/- 3.26, N = 3 SE +/- 2.86, N = 6 SE +/- 3.10, N = 5 337.32 337.44 562.50 529.80 597.73 528.21 450.12 420.35 465.06 499.18 474.68 467.96 486.05 507.56 455.79 443.58 MIN: 35.76 / MAX: 4000 MIN: 33.44 / MAX: 4615.38 MIN: 57.47 / MAX: 6666.67 MIN: 86.08 / MAX: 5454.55 MIN: 61.41 / MAX: 5454.55 MIN: 92.17 / MAX: 6000 MIN: 66.15 / MAX: 5454.55 MIN: 78.33 / MAX: 4285.71 MIN: 32.59 / MAX: 5000 MIN: 54.2 / MAX: 4615.38 MIN: 36.43 / MAX: 6000 MIN: 52.36 / MAX: 6000 MIN: 40.43 / MAX: 6666.67 MIN: 66.45 / MAX: 5454.55 MIN: 42.19 / MAX: 6666.67 MIN: 62.43 / MAX: 6666.67
NWChem NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.
OpenRadioss OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/ and https://github.com/OpenRadioss/ModelExchange/tree/main/Examples. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Cell Phone Drop Test EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8 16 24 32 40 SE +/- 0.18, N = 3 SE +/- 0.35, N = 3 SE +/- 0.26, N = 3 SE +/- 0.05, N = 3 SE +/- 0.19, N = 3 SE +/- 0.05, N = 3 SE +/- 0.26, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 SE +/- 0.18, N = 3 SE +/- 0.32, N = 3 33.23 35.95 22.53 23.53 20.45 22.52 23.69 27.86 33.75 27.51 31.92 27.63 31.49 28.22 27.90 26.88 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenRadioss 2023.09.15 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 19.07 / Avg: 127.55 / Max: 161.94 Min: 18.07 / Avg: 113.05 / Max: 140.62 Min: 30.24 / Avg: 238.37 / Max: 343.81 Min: 58.09 / Avg: 439.02 / Max: 706.97 Min: 33.62 / Avg: 276.69 / Max: 409.57 Min: 60.04 / Avg: 493.1 / Max: 819.11 Min: 19.86 / Avg: 210.23 / Max: 333.55 Min: 41.04 / Avg: 357.25 / Max: 670.96 Min: 76.63 / Avg: 302.43 / Max: 377 Min: 199.37 / Avg: 570.84 / Max: 753.68 Min: 77.53 / Avg: 297.92 / Max: 375.01 Min: 206.26 / Avg: 572.24 / Max: 749.99 Min: 59.2 / Avg: 289.5 / Max: 374.11 Min: 172.03 / Avg: 545.45 / Max: 745.98 Min: 34.28 / Avg: 272.03 / Max: 374.83 Min: 71.17 / Avg: 504.55 / Max: 754.24
ClickHouse ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 SE +/- 1.57, N = 3 SE +/- 1.88, N = 3 SE +/- 5.33, N = 3 SE +/- 6.05, N = 3 SE +/- 7.88, N = 3 SE +/- 4.34, N = 3 SE +/- 2.78, N = 3 SE +/- 4.49, N = 5 SE +/- 2.85, N = 3 SE +/- 1.35, N = 3 SE +/- 5.88, N = 3 SE +/- 3.85, N = 12 SE +/- 3.41, N = 3 SE +/- 4.37, N = 3 SE +/- 4.18, N = 6 SE +/- 4.12, N = 5 323.59 322.69 536.82 504.84 566.31 513.92 433.79 411.26 446.40 491.89 461.62 460.66 465.50 485.63 437.20 423.30 MIN: 35.69 / MAX: 3750 MIN: 32.99 / MAX: 4000 MIN: 56.6 / MAX: 6666.67 MIN: 81.97 / MAX: 4615.38 MIN: 59.64 / MAX: 5454.55 MIN: 84.87 / MAX: 5000 MIN: 54.1 / MAX: 4285.71 MIN: 77.12 / MAX: 5000 MIN: 32.64 / MAX: 4615.38 MIN: 51.86 / MAX: 5454.55 MIN: 35.59 / MAX: 5454.55 MIN: 53.62 / MAX: 5454.55 MIN: 40.93 / MAX: 5454.55 MIN: 63.63 / MAX: 5000 MIN: 41.7 / MAX: 5454.55 MIN: 61.54 / MAX: 6000
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.66, N = 3 SE +/- 0.57, N = 3 SE +/- 1.39, N = 4 SE +/- 0.50, N = 3 SE +/- 2.00, N = 12 SE +/- 0.18, N = 3 SE +/- 0.14, N = 3 SE +/- 1.40, N = 3 SE +/- 0.63, N = 3 SE +/- 1.04, N = 12 SE +/- 0.03, N = 3 SE +/- 1.57, N = 3 SE +/- 0.58, N = 3 SE +/- 0.21, N = 3 98.70 92.82 138.11 133.87 123.37 120.11 125.76 131.38 119.73 141.95 123.69 135.28 143.67 161.06 129.02 150.33 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.1555 0.311 0.4665 0.622 0.7775 0.646 0.691 0.485 0.252 0.385 0.203 0.525 0.291 0.354 0.210 0.363 0.200 0.425 0.243 0.463 0.278 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.16.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 18.72 / Avg: 152.81 / Max: 164.9 Min: 16.01 / Avg: 134.32 / Max: 142.38 Min: 29.3 / Avg: 284.71 / Max: 314.17 Min: 53.87 / Avg: 531.98 / Max: 578.15 Min: 33.82 / Avg: 320.09 / Max: 348.64 Min: 58.73 / Avg: 590.53 / Max: 641.68 Min: 20.95 / Avg: 239.33 / Max: 288.65 Min: 43.33 / Avg: 451.33 / Max: 504.94 Min: 73.46 / Avg: 338.38 / Max: 359.94 Min: 203.91 / Avg: 674.94 / Max: 716.61 Min: 77.09 / Avg: 340.59 / Max: 361.18 Min: 196.92 / Avg: 674.95 / Max: 729.37 Min: 59.06 / Avg: 338.33 / Max: 360.73 Min: 163.23 / Avg: 663.98 / Max: 713.39 Min: 33.15 / Avg: 278.36 / Max: 312.29 Min: 84.37 / Avg: 540.26 / Max: 604.56
Timed Godot Game Engine Compilation This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 4.0 Time To Compile EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 SE +/- 0.32, N = 3 SE +/- 0.16, N = 3 SE +/- 0.16, N = 3 SE +/- 0.06, N = 3 SE +/- 0.12, N = 3 SE +/- 0.62, N = 3 SE +/- 0.20, N = 3 SE +/- 0.55, N = 3 SE +/- 0.18, N = 3 SE +/- 0.56, N = 3 SE +/- 0.32, N = 3 SE +/- 1.69, N = 3 SE +/- 0.76, N = 3 SE +/- 0.56, N = 3 SE +/- 0.91, N = 3 SE +/- 0.29, N = 3 154.01 159.62 111.69 102.58 103.35 96.16 128.28 126.11 149.38 123.88 143.81 122.70 140.23 118.57 122.72 103.38 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed Godot Game Engine Compilation 4.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.58 / Avg: 110.14 / Max: 169.2 Min: 16.76 / Avg: 97.75 / Max: 142.01 Min: 24.79 / Avg: 188.63 / Max: 336.96 Min: 58.53 / Avg: 338.4 / Max: 689.21 Min: 33.53 / Avg: 222.04 / Max: 410.61 Min: 65.36 / Avg: 383.39 / Max: 820.55 Min: 18.92 / Avg: 152.32 / Max: 333.75 Min: 39.69 / Avg: 276.98 / Max: 673.64 Min: 76.87 / Avg: 285.63 / Max: 375.65 Min: 199.22 / Avg: 527.95 / Max: 724.51 Min: 82.82 / Avg: 282.51 / Max: 375.66 Min: 197.93 / Avg: 519.6 / Max: 744.47 Min: 55.69 / Avg: 255.97 / Max: 375.31 Min: 154.59 / Avg: 461.39 / Max: 752.04 Min: 33.4 / Avg: 236.2 / Max: 374.96 Min: 82.98 / Avg: 430.41 / Max: 750.75
QMCPACK QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. QMCPACK is an open-source production level many-body ab initio Quantum Monte Carlo code for computing the electronic structure of atoms, molecules, and solids. QMCPACK is supported by the U.S. Department of Energy. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.17.1 Input: Li2_STO_ae EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 30 60 90 120 150 SE +/- 1.13, N = 3 SE +/- 0.40, N = 3 SE +/- 0.54, N = 3 SE +/- 0.23, N = 3 SE +/- 1.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.63, N = 3 SE +/- 0.42, N = 3 SE +/- 1.20, N = 3 SE +/- 0.88, N = 3 SE +/- 1.04, N = 3 SE +/- 1.11, N = 5 SE +/- 0.89, N = 3 SE +/- 1.11, N = 3 SE +/- 1.03, N = 3 SE +/- 0.93, N = 3 137.23 151.27 108.02 102.98 97.00 91.97 110.80 111.15 121.19 102.12 117.81 106.27 109.30 101.63 105.98 97.91 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better QMCPACK 3.17.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.96 / Avg: 154.43 / Max: 159.92 Min: 17.04 / Avg: 132.24 / Max: 137.56 Min: 28.38 / Avg: 316.1 / Max: 334.43 Min: 54.02 / Avg: 635.44 / Max: 688.42 Min: 33.88 / Avg: 391.36 / Max: 409.49 Min: 66.83 / Avg: 767.28 / Max: 818.97 Min: 21.99 / Avg: 319.33 / Max: 338.99 Min: 44.37 / Avg: 628.81 / Max: 690.2 Min: 75.52 / Avg: 323.64 / Max: 364.79 Min: 208.9 / Avg: 673.9 / Max: 748.46 Min: 80.16 / Avg: 338.77 / Max: 371.37 Min: 206.85 / Avg: 676.4 / Max: 753.87 Min: 59.34 / Avg: 337.54 / Max: 376.68 Min: 169.74 / Avg: 670.52 / Max: 753.5 Min: 23.78 / Avg: 335.13 / Max: 377.2 Min: 63.82 / Avg: 660.86 / Max: 753.6
Timed FFmpeg Compilation This test times how long it takes to build the FFmpeg multimedia library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed FFmpeg Compilation 7.0 Time To Compile EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 7 14 21 28 35 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.30, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 27.54 28.42 20.49 18.09 19.99 17.84 23.22 21.22 26.46 20.79 24.84 20.01 24.06 19.51 21.36 17.39 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed FFmpeg Compilation 7.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.56 / Avg: 90.17 / Max: 171.06 Min: 17.2 / Avg: 81.55 / Max: 145.14 Min: 28.44 / Avg: 151.49 / Max: 345.46 Min: 55.22 / Avg: 281.94 / Max: 701.16 Min: 32.76 / Avg: 171.98 / Max: 410.22 Min: 65.37 / Avg: 314.5 / Max: 819.41 Min: 21.46 / Avg: 124.42 / Max: 336.98 Min: 39.64 / Avg: 238.4 / Max: 679.1 Min: 74.35 / Avg: 250.86 / Max: 374.99 Min: 208.48 / Avg: 470.97 / Max: 753.46 Min: 81.57 / Avg: 251.37 / Max: 376.37 Min: 198.1 / Avg: 467.2 / Max: 752.99 Min: 57.21 / Avg: 220.08 / Max: 376.48 Min: 157.82 / Avg: 400.67 / Max: 753.15 Min: 34.1 / Avg: 198.27 / Max: 377.58 Min: 85.35 / Avg: 348.01 / Max: 755.27
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 163.70 188.37 136.40 140.48 118.54 135.52 167.41 192.42 153.12 134.78 150.67 159.31 160.73 162.03 148.40 139.62 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.43 Operation: Noise-Gaussian EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 SE +/- 0.00, N = 3 SE +/- 0.67, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 1.15, N = 3 SE +/- 0.00, N = 3 SE +/- 0.88, N = 3 123 120 160 178 165 179 151 164 150 150 158 164 158 182 171 194 1. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp Iterations Per Minute Per Watt
OpenBenchmarking.org Iterations Per Minute Per Watt, More Is Better GraphicsMagick 1.3.43 Operation: Noise-Gaussian EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.3391 0.6782 1.0173 1.3564 1.6955 1.420 1.507 1.041 0.597 0.932 0.540 1.226 0.674 0.552 0.308 0.614 0.345 0.716 0.467 0.778 0.544 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better GraphicsMagick 1.3.43 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90 180 270 360 450 Min: 17.9 / Avg: 86.65 / Max: 95.02 Min: 16.29 / Avg: 79.6 / Max: 84.69 Min: 29.47 / Avg: 153.63 / Max: 171.14 Min: 59.18 / Avg: 297.92 / Max: 323.55 Min: 33.85 / Avg: 176.95 / Max: 194.06 Min: 66.05 / Avg: 331.54 / Max: 359.49 Min: 21.53 / Avg: 123.19 / Max: 140.7 Min: 39.77 / Avg: 243.28 / Max: 272.18 Min: 74.91 / Avg: 271.55 / Max: 302.98 Min: 203.35 / Avg: 486.24 / Max: 527.71 Min: 75.67 / Avg: 257.27 / Max: 283.18 Min: 196.74 / Avg: 476 / Max: 507.38 Min: 59.26 / Avg: 220.76 / Max: 250.24 Min: 160.72 / Avg: 389.84 / Max: 423.25 Min: 34.53 / Avg: 219.79 / Max: 241.79 Min: 82.86 / Avg: 356.85 / Max: 403.74
SVT-AV1 This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.0 Encoder Mode: Preset 8 - Input: Bosphorus 4K EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20 40 60 80 100 SE +/- 0.11, N = 4 SE +/- 0.67, N = 4 SE +/- 0.56, N = 5 SE +/- 0.40, N = 5 SE +/- 0.42, N = 5 SE +/- 0.63, N = 5 SE +/- 0.38, N = 5 SE +/- 0.31, N = 4 SE +/- 0.26, N = 4 SE +/- 0.58, N = 5 SE +/- 0.29, N = 4 SE +/- 0.44, N = 15 SE +/- 0.20, N = 5 SE +/- 0.46, N = 4 SE +/- 0.22, N = 5 SE +/- 0.79, N = 5 68.19 64.79 87.99 86.34 86.30 87.82 78.74 77.60 62.75 54.82 65.36 55.98 70.91 65.22 79.44 70.43 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 2.0 Encoder Mode: Preset 8 - Input: Bosphorus 4K EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.2147 0.4294 0.6441 0.8588 1.0735 0.879 0.954 0.653 0.349 0.570 0.321 0.737 0.388 0.278 0.124 0.286 0.126 0.362 0.179 0.439 0.227 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 2.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 100 200 300 400 500 Min: 16.05 / Avg: 77.57 / Max: 129.87 Min: 14.98 / Avg: 67.93 / Max: 107.89 Min: 28.36 / Avg: 134.72 / Max: 238.02 Min: 56.85 / Avg: 247.1 / Max: 379.6 Min: 31.53 / Avg: 151.3 / Max: 266.73 Min: 62.22 / Avg: 273.59 / Max: 425.17 Min: 21.17 / Avg: 106.8 / Max: 187.26 Min: 39.92 / Avg: 200.17 / Max: 302.35 Min: 76.76 / Avg: 226.08 / Max: 319.38 Min: 193.1 / Avg: 442.35 / Max: 576.63 Min: 73.57 / Avg: 228.42 / Max: 325.25 Min: 196.2 / Avg: 442.62 / Max: 580.63 Min: 59.76 / Avg: 195.92 / Max: 293.59 Min: 171.54 / Avg: 364.29 / Max: 488.45 Min: 27.5 / Avg: 181.13 / Max: 316.97 Min: 70.26 / Avg: 309.78 / Max: 476.6
Timed Gem5 Compilation This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Gem5 Compilation 23.0.1 Time To Compile EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 1.99, N = 9 SE +/- 0.97, N = 3 SE +/- 2.24, N = 4 SE +/- 0.72, N = 3 SE +/- 2.15, N = 3 SE +/- 1.16, N = 3 SE +/- 1.88, N = 3 SE +/- 1.73, N = 3 SE +/- 2.59, N = 4 SE +/- 1.72, N = 3 SE +/- 2.11, N = 6 SE +/- 1.01, N = 3 SE +/- 0.92, N = 3 SE +/- 0.58, N = 3 SE +/- 2.17, N = 3 SE +/- 0.89, N = 3 242.93 251.61 185.30 164.90 176.88 161.94 212.55 192.98 220.57 184.30 215.50 181.63 210.93 178.28 187.51 157.36 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed Gem5 Compilation 23.0.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.32 / Avg: 96.65 / Max: 166.6 Min: 16.66 / Avg: 87.52 / Max: 141.58 Min: 29.61 / Avg: 170.81 / Max: 332.22 Min: 55.44 / Avg: 320.45 / Max: 676.84 Min: 29.1 / Avg: 201.66 / Max: 413.09 Min: 65.57 / Avg: 360.36 / Max: 822.7 Min: 21.18 / Avg: 144.58 / Max: 331.44 Min: 40.43 / Avg: 270.24 / Max: 677.31 Min: 40.79 / Avg: 257.31 / Max: 376.35 Min: 177.84 / Avg: 479.92 / Max: 746.22 Min: 67.72 / Avg: 254.21 / Max: 375.35 Min: 191.44 / Avg: 472.13 / Max: 747.43 Min: 55.65 / Avg: 228.11 / Max: 376.67 Min: 160.34 / Avg: 415.56 / Max: 738.65 Min: 33.1 / Avg: 215.42 / Max: 377.33 Min: 43.96 / Avg: 377.81 / Max: 753.32
Timed ImageMagick Compilation This test times how long it takes to build ImageMagick. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 4 SE +/- 0.04, N = 4 SE +/- 0.06, N = 4 SE +/- 0.04, N = 4 SE +/- 0.08, N = 4 SE +/- 0.02, N = 4 SE +/- 0.15, N = 4 SE +/- 0.11, N = 4 SE +/- 0.06, N = 4 SE +/- 0.07, N = 4 SE +/- 0.10, N = 4 SE +/- 0.13, N = 4 SE +/- 0.10, N = 4 SE +/- 0.03, N = 4 17.92 18.18 13.47 12.75 12.20 11.72 16.39 15.81 16.08 14.40 15.38 14.09 14.72 13.26 12.60 11.49 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed ImageMagick Compilation 6.9.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.83 / Avg: 69.48 / Max: 164.08 Min: 16.43 / Avg: 65.12 / Max: 142.84 Min: 28.39 / Avg: 120.39 / Max: 332.54 Min: 57.57 / Avg: 245.77 / Max: 578.82 Min: 32.48 / Avg: 139.23 / Max: 409.7 Min: 62.42 / Avg: 273.56 / Max: 669.18 Min: 21.44 / Avg: 97.43 / Max: 329.22 Min: 40.62 / Avg: 204.88 / Max: 561.52 Min: 73.52 / Avg: 220.79 / Max: 376.49 Min: 203.42 / Avg: 435.59 / Max: 750.99 Min: 75.62 / Avg: 220.24 / Max: 379.25 Min: 198.51 / Avg: 429.23 / Max: 750.07 Min: 60.5 / Avg: 186.22 / Max: 378.44 Min: 164.76 / Avg: 362.03 / Max: 752.14 Min: 26.01 / Avg: 162.46 / Max: 377 Min: 81.81 / Avg: 295.33 / Max: 748.04
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 512 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200M 400M 600M 800M 1000M SE +/- 822739.33, N = 3 SE +/- 2292977.20, N = 3 SE +/- 1678355.68, N = 3 SE +/- 2261241.06, N = 3 SE +/- 752558.60, N = 3 SE +/- 3039034.93, N = 3 SE +/- 673003.22, N = 3 SE +/- 958772.36, N = 3 SE +/- 1611276.24, N = 3 SE +/- 4432294.36, N = 3 SE +/- 1289603.73, N = 3 SE +/- 6141471.96, N = 3 SE +/- 940265.92, N = 3 SE +/- 3717554.45, N = 3 SE +/- 1515552.70, N = 3 SE +/- 7562646.32, N = 8 633860000 577263333 811373333 824583333 850586667 849340000 709650000 709393333 632396667 732560000 689003333 757176667 744410000 799496667 783770000 883701250 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 512 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1.2M 2.4M 3.6M 4.8M 6M 5046791.28 5458880.22 3317050.16 2172222.16 2753902.96 1904459.88 3748079.60 2394875.44 1991130.78 1151612.89 2174337.88 1184768.10 2362548.08 1519282.71 2577646.65 1705499.35 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.47 / Avg: 125.6 / Max: 142.49 Min: 16.94 / Avg: 105.75 / Max: 120.76 Min: 28.29 / Avg: 244.61 / Max: 277.01 Min: 59.09 / Avg: 379.6 / Max: 439.61 Min: 34.73 / Avg: 308.87 / Max: 353.22 Min: 65.79 / Avg: 445.97 / Max: 506.44 Min: 18.89 / Avg: 189.34 / Max: 213.55 Min: 39.32 / Avg: 296.21 / Max: 332.18 Min: 79.61 / Avg: 317.61 / Max: 374.24 Min: 195.75 / Avg: 636.12 / Max: 707.28 Min: 80.39 / Avg: 316.88 / Max: 376.81 Min: 148.91 / Avg: 639.09 / Max: 734.49 Min: 61.24 / Avg: 315.09 / Max: 375.37 Min: 164.45 / Avg: 526.23 / Max: 645.98 Min: 26.5 / Avg: 304.06 / Max: 370.78 Min: 67.01 / Avg: 518.15 / Max: 711.65
Appleseed Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.
Scene: Disney Material
EPYC 9654 2P: The test run did not produce a result.
EPYC 9684X 2P: The test run did not produce a result.
EPYC 9754 2P: The test run did not produce a result.
Zstd Compression This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Compression Speed EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 15 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 6.76 6.74 8.38 8.30 8.45 8.42 6.68 6.64 7.76 7.62 7.72 7.46 7.94 8.16 9.80 9.67 1. (CC) gcc options: -O3 -pthread -lz -llzma
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2024.0 Model: Face Detection Retail FP16-INT8 - Device: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 6.77 7.60 5.85 5.98 5.41 5.34 6.25 6.60 5.27 5.55 5.77 6.02 6.43 5.72 5.19 5.18 MIN: 3.54 / MAX: 18.38 MIN: 3.84 / MAX: 21.81 MIN: 2.67 / MAX: 22.66 MIN: 4.9 / MAX: 26.69 MIN: 2.65 / MAX: 24.3 MIN: 4.88 / MAX: 26.62 MIN: 3.07 / MAX: 31.02 MIN: 5.84 / MAX: 35.85 MIN: 4.98 / MAX: 13.16 MIN: 5.12 / MAX: 24.99 MIN: 5.26 / MAX: 13.86 MIN: 5.37 / MAX: 27.77 MIN: 3.84 / MAX: 15.83 MIN: 5.1 / MAX: 17.93 MIN: 4.72 / MAX: 11.02 MIN: 4.74 / MAX: 26.35 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better OpenVINO 2024.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.15 / Avg: 159.45 / Max: 171.38 Min: 16.94 / Avg: 130.25 / Max: 140.67 Min: 31.29 / Avg: 316.48 / Max: 340.73 Min: 59.24 / Avg: 612.41 / Max: 670.98 Min: 30.34 / Avg: 377.27 / Max: 406.75 Min: 65.72 / Avg: 754.92 / Max: 818.94 Min: 21 / Avg: 311.43 / Max: 339.91 Min: 39.66 / Avg: 604.27 / Max: 668.8 Min: 72.13 / Avg: 330.52 / Max: 368.64 Min: 196.21 / Avg: 665.03 / Max: 748.66 Min: 77.72 / Avg: 331.34 / Max: 373.94 Min: 203.86 / Avg: 663.32 / Max: 752.94 Min: 58.3 / Avg: 329.39 / Max: 378.11 Min: 165.34 / Avg: 658.39 / Max: 751.53 Min: 35.16 / Avg: 326.91 / Max: 374.78 Min: 83.99 / Avg: 648.24 / Max: 768.1
Zstd Compression This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300 600 900 1200 1500 SE +/- 21.23, N = 3 SE +/- 1.58, N = 3 SE +/- 0.46, N = 3 SE +/- 2.20, N = 3 SE +/- 7.31, N = 3 SE +/- 1.73, N = 3 SE +/- 0.83, N = 3 SE +/- 1.83, N = 3 SE +/- 2.12, N = 3 SE +/- 0.66, N = 15 SE +/- 1.08, N = 15 SE +/- 0.65, N = 15 SE +/- 4.45, N = 3 SE +/- 4.96, N = 3 SE +/- 3.01, N = 4 SE +/- 1.88, N = 3 1118.8 1144.8 1388.9 1382.1 1385.3 1392.5 1139.9 1139.0 1126.4 1040.2 1127.5 1039.2 1032.0 963.6 1221.8 1165.8 1. (CC) gcc options: -O3 -pthread -lz -llzma MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 19.698 20.546 13.308 5.818 11.711 5.292 14.065 5.875 5.616 2.506 5.548 2.520 6.465 2.828 8.705 4.008 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Zstd Compression 1.5.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 80 160 240 320 400 Min: 16.44 / Avg: 56.8 / Max: 67.51 Min: 16.36 / Avg: 55.72 / Max: 65.47 Min: 27.93 / Avg: 104.36 / Max: 120.51 Min: 59.25 / Avg: 237.56 / Max: 259.24 Min: 33.68 / Avg: 118.29 / Max: 136.36 Min: 60.19 / Avg: 263.13 / Max: 287.32 Min: 21.5 / Avg: 81.05 / Max: 92.77 Min: 38.29 / Avg: 193.87 / Max: 209.99 Min: 73.69 / Avg: 200.55 / Max: 225.84 Min: 203.74 / Avg: 415.14 / Max: 451.39 Min: 75.37 / Avg: 203.22 / Max: 228.29 Min: 197.43 / Avg: 412.33 / Max: 448.69 Min: 61.06 / Avg: 159.64 / Max: 183.39 Min: 170.25 / Avg: 340.73 / Max: 371.6 Min: 33.62 / Avg: 140.35 / Max: 172.45 Min: 76.44 / Avg: 290.84 / Max: 334.48
VVenC VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Faster EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.13, N = 3 SE +/- 0.13, N = 3 SE +/- 0.02, N = 3 SE +/- 0.16, N = 3 SE +/- 0.08, N = 3 SE +/- 0.12, N = 3 13.97 13.61 16.73 17.02 14.79 14.10 11.90 14.36 11.97 15.06 12.54 15.97 13.26 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Faster EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0322 0.0644 0.0966 0.1288 0.161 0.130 0.143 0.090 0.080 0.106 0.051 0.023 0.051 0.023 0.060 0.028 0.068 0.031 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better VVenC 1.11 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 Min: 18.52 / Avg: 107.71 / Max: 147.66 Min: 17.01 / Avg: 95.43 / Max: 126.76 Min: 29.27 / Avg: 185.29 / Max: 257.81 Min: 34.49 / Avg: 211.98 / Max: 303.77 Min: 20.86 / Avg: 139.36 / Max: 195.74 Min: 74.46 / Avg: 275.77 / Max: 340.06 Min: 209.37 / Avg: 521.79 / Max: 619.75 Min: 73.18 / Avg: 282.13 / Max: 350.27 Min: 204.58 / Avg: 521.13 / Max: 631.04 Min: 60.17 / Avg: 250.97 / Max: 326.74 Min: 149.66 / Avg: 449.44 / Max: 572.09 Min: 34.91 / Avg: 235.39 / Max: 320.85 Min: 77.34 / Avg: 425.45 / Max: 535.49 Video Input: Bosphorus 4K - Video Preset: Faster
EPYC 9654 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1
EPYC 9684X 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1
EPYC 9754 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8 16 24 32 40 33.67 36.82 28.03 27.74 25.93 26.82 30.38 29.47 31.78 32.48 32.00 28.84 31.38 35.60 31.70 29.33 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200 400 600 800 1000 SE +/- 1.45, N = 3 SE +/- 2.40, N = 3 SE +/- 2.19, N = 3 SE +/- 1.86, N = 3 SE +/- 2.08, N = 3 SE +/- 1.20, N = 3 SE +/- 0.58, N = 3 SE +/- 1.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.67, N = 3 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 SE +/- 1.15, N = 3 SE +/- 0.41, N = 4 SE +/- 0.48, N = 4 976 974 818 822 818 818 974 972 763 763 762 763 765 767 691 692 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better PyBench 2018-02-16 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 60 120 180 240 300 Min: 15.93 / Avg: 48.41 / Max: 54.07 Min: 14.86 / Avg: 47.57 / Max: 53.35 Min: 28.15 / Avg: 88.41 / Max: 100.93 Min: 57.47 / Avg: 207.73 / Max: 239.74 Min: 33.11 / Avg: 99.97 / Max: 113.9 Min: 65.07 / Avg: 232.2 / Max: 264.86 Min: 19.44 / Avg: 72.53 / Max: 81.29 Min: 38.38 / Avg: 175.13 / Max: 196.99 Min: 74.61 / Avg: 114.52 / Max: 125.95 Min: 206.86 / Avg: 276.72 / Max: 343.62 Min: 83.03 / Avg: 121.48 / Max: 133.99 Min: 205.54 / Avg: 276.95 / Max: 327.1 Min: 58.93 / Avg: 88.45 / Max: 123.25 Min: 163.51 / Avg: 228.16 / Max: 305.07 Min: 33.53 / Avg: 85.96 / Max: 134.17 Min: 79.78 / Avg: 194.14 / Max: 289.01
Liquid-DSP LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 57 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 600M 1200M 1800M 2400M 3000M SE +/- 4914717.13, N = 3 SE +/- 7776531.65, N = 3 SE +/- 3786085.64, N = 3 SE +/- 10590929.03, N = 3 SE +/- 8476044.68, N = 3 SE +/- 8732315.46, N = 3 SE +/- 8567444.84, N = 3 SE +/- 17457503.96, N = 3 SE +/- 2629533.12, N = 3 SE +/- 28812627.44, N = 4 SE +/- 5493127.02, N = 3 SE +/- 27382304.50, N = 5 SE +/- 7076957.92, N = 3 SE +/- 28500058.48, N = 3 SE +/- 11426868.92, N = 3 SE +/- 32405829.38, N = 3 2186833333 2026333333 2859833333 2730233333 2862000000 2772800000 2444933333 2374766667 2138766667 2494050000 2308966667 2491340000 2539200000 2655500000 2475500000 2620066667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 57 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4M 8M 12M 16M 20M 16817799.69 18884627.20 10813033.93 6982683.34 9250164.53 6245758.05 12777153.28 8048978.36 6748464.67 3917514.64 7252530.07 3870621.21 8050132.03 4994066.82 8044246.29 5119414.53 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.61 / Avg: 130.03 / Max: 147.65 Min: 16.16 / Avg: 107.3 / Max: 126.15 Min: 27.79 / Avg: 264.48 / Max: 300.86 Min: 54.92 / Avg: 391 / Max: 447.88 Min: 32.57 / Avg: 309.4 / Max: 348.95 Min: 64 / Avg: 443.95 / Max: 498.34 Min: 18.99 / Avg: 191.35 / Max: 215.83 Min: 37.21 / Avg: 295.04 / Max: 331.11 Min: 74.72 / Avg: 316.93 / Max: 377.16 Min: 203.4 / Avg: 636.64 / Max: 709.85 Min: 76.69 / Avg: 318.37 / Max: 376.42 Min: 206.33 / Avg: 643.65 / Max: 732.46 Min: 60.17 / Avg: 315.42 / Max: 375.58 Min: 154.62 / Avg: 531.73 / Max: 660.07 Min: 27.41 / Avg: 307.74 / Max: 373.74 Min: 67.62 / Avg: 511.79 / Max: 675.09
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 32 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 500M 1000M 1500M 2000M 2500M SE +/- 3090487.20, N = 3 SE +/- 3256958.36, N = 3 SE +/- 2810891.52, N = 3 SE +/- 1567730.14, N = 3 SE +/- 3000185.18, N = 3 SE +/- 2510644.01, N = 3 SE +/- 1550627.11, N = 3 SE +/- 1462114.15, N = 3 SE +/- 688799.28, N = 3 SE +/- 2206304.10, N = 3 SE +/- 692820.32, N = 3 SE +/- 3017909.50, N = 3 SE +/- 3167192.94, N = 3 SE +/- 3300505.01, N = 3 SE +/- 27191440.65, N = 3 SE +/- 17236814.61, N = 3 1871333333 1776866667 2282766667 2282066667 2287466667 2279300000 1898766667 1894966667 1640266667 1963233333 1679700000 2090466667 1890566667 2025000000 1927766667 2154633333 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 32 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4M 8M 12M 16M 20M 14706486.76 17627091.46 9937920.60 6304024.25 8884551.10 5779445.49 11984070.91 7114576.18 6685402.04 3308597.89 6609424.93 3387722.73 6722664.60 3966847.18 6720931.25 4527054.56 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 Min: 15.76 / Avg: 127.25 / Max: 146.72 Min: 15.81 / Avg: 100.8 / Max: 114.74 Min: 25.11 / Avg: 229.7 / Max: 257.97 Min: 57.58 / Avg: 362 / Max: 403.93 Min: 33.54 / Avg: 257.47 / Max: 292.35 Min: 64.31 / Avg: 394.38 / Max: 445.2 Min: 20.46 / Avg: 158.44 / Max: 178.44 Min: 38.6 / Avg: 266.35 / Max: 296.68 Min: 76.61 / Avg: 245.35 / Max: 268.44 Min: 200.67 / Avg: 593.37 / Max: 651.17 Min: 74.71 / Avg: 254.14 / Max: 326.8 Min: 197.41 / Avg: 617.07 / Max: 678.02 Min: 60.85 / Avg: 281.22 / Max: 356.43 Min: 152.44 / Avg: 510.48 / Max: 610.23 Min: 27.76 / Avg: 286.83 / Max: 364.52 Min: 75.17 / Avg: 475.95 / Max: 586.43
QuantLib QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Single-Threaded EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 800 1600 2400 3200 4000 SE +/- 18.64, N = 3 SE +/- 14.40, N = 3 SE +/- 24.27, N = 3 SE +/- 19.83, N = 3 SE +/- 22.39, N = 3 SE +/- 1.14, N = 3 SE +/- 14.83, N = 3 SE +/- 19.30, N = 3 SE +/- 22.17, N = 3 SE +/- 28.11, N = 3 SE +/- 25.35, N = 3 SE +/- 28.57, N = 3 SE +/- 28.42, N = 3 SE +/- 25.40, N = 3 SE +/- 33.58, N = 3 SE +/- 30.36, N = 12 2593.3 2593.0 3105.9 3098.5 3114.2 3125.4 2596.1 2590.8 3225.7 3207.0 3229.1 3222.8 3209.6 3210.1 3600.5 3586.3 1. (CXX) g++ options: -O3 -march=native -fPIE -pie MFLOPS Per Watt
OpenBenchmarking.org MFLOPS Per Watt, More Is Better QuantLib 1.32 Configuration: Single-Threaded EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 11 22 33 44 55 49.12 50.14 33.74 14.23 29.60 12.80 34.81 14.07 26.68 11.15 24.92 11.01 31.94 13.29 45.35 15.01 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better QuantLib 1.32 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 Min: 17.54 / Avg: 52.79 / Max: 57.91 Min: 17.21 / Avg: 51.72 / Max: 57.08 Min: 29.32 / Avg: 92.05 / Max: 101.3 Min: 57.57 / Avg: 217.74 / Max: 241.62 Min: 32.36 / Avg: 105.22 / Max: 115.88 Min: 66.1 / Avg: 244.1 / Max: 269.19 Min: 20.77 / Avg: 74.58 / Max: 81.84 Min: 40.16 / Avg: 184.09 / Max: 202.33 Min: 69.08 / Avg: 120.93 / Max: 183.43 Min: 202.91 / Avg: 287.73 / Max: 404.53 Min: 80.8 / Avg: 129.58 / Max: 191.82 Min: 208.95 / Avg: 292.75 / Max: 399.59 Min: 55.04 / Avg: 100.48 / Max: 155.72 Min: 161.45 / Avg: 241.52 / Max: 334.35 Min: 31.58 / Avg: 79.4 / Max: 103.91 Min: 75.25 / Avg: 238.94 / Max: 292.83
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.17 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8 16 24 32 40 SE +/- 0.14, N = 3 SE +/- 0.25, N = 3 SE +/- 0.36, N = 3 SE +/- 0.05, N = 3 SE +/- 0.19, N = 3 SE +/- 0.08, N = 3 SE +/- 0.33, N = 4 SE +/- 0.24, N = 7 SE +/- 0.10, N = 3 SE +/- 0.34, N = 3 SE +/- 0.07, N = 3 SE +/- 0.19, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 27.94 27.87 29.84 25.63 28.73 24.05 27.45 26.33 28.52 24.73 28.48 24.68 32.34 30.04 33.38 31.71 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 9 18 27 36 45 SE +/- 0.18, N = 3 SE +/- 0.33, N = 3 SE +/- 0.41, N = 3 SE +/- 0.07, N = 3 SE +/- 0.22, N = 3 SE +/- 0.15, N = 3 SE +/- 0.43, N = 4 SE +/- 0.35, N = 7 SE +/- 0.13, N = 3 SE +/- 0.55, N = 3 SE +/- 0.08, N = 3 SE +/- 0.32, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 35.79 35.88 33.52 39.02 34.81 41.57 36.45 38.00 35.07 40.45 35.11 40.52 30.92 33.29 29.96 31.53 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
srsRAN Project srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mbps, More Is Better srsRAN Project 23.10.1-20240325 Test: PDSCH Processor Benchmark, Throughput Thread EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 160 320 480 640 800 SE +/- 3.14, N = 7 SE +/- 1.24, N = 7 SE +/- 1.71, N = 8 SE +/- 3.72, N = 8 SE +/- 3.13, N = 8 SE +/- 5.52, N = 8 SE +/- 0.80, N = 7 SE +/- 4.48, N = 7 SE +/- 1.64, N = 8 SE +/- 3.86, N = 8 SE +/- 1.27, N = 8 SE +/- 5.02, N = 8 SE +/- 1.17, N = 8 SE +/- 2.57, N = 8 SE +/- 1.34, N = 8 SE +/- 1.89, N = 8 557.9 563.6 684.7 673.4 680.7 664.7 596.2 576.8 702.6 651.9 705.5 663.8 680.2 675.5 742.9 755.5 1. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl Mbps Per Watt
OpenBenchmarking.org Mbps Per Watt, More Is Better srsRAN Project 23.10.1-20240325 Test: PDSCH Processor Benchmark, Throughput Thread EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4 8 12 16 20 13.548 14.524 9.795 4.162 8.628 3.639 10.675 4.200 5.085 2.035 4.788 2.065 6.116 2.598 8.929 4.074 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 Min: 18.39 / Avg: 41.18 / Max: 58.06 Min: 13.69 / Avg: 38.81 / Max: 55.52 Min: 28.33 / Avg: 69.9 / Max: 103.51 Min: 56.59 / Avg: 161.78 / Max: 243.41 Min: 30.33 / Avg: 78.89 / Max: 117.61 Min: 63.99 / Avg: 182.65 / Max: 268.9 Min: 20.14 / Avg: 55.85 / Max: 81.57 Min: 38.06 / Avg: 137.34 / Max: 200.95 Min: 73.91 / Avg: 138.16 / Max: 195.36 Min: 204.49 / Avg: 320.28 / Max: 415.75 Min: 85.93 / Avg: 147.34 / Max: 200.38 Min: 199.86 / Avg: 321.38 / Max: 411.1 Min: 62.63 / Avg: 111.22 / Max: 152.68 Min: 161.37 / Avg: 259.97 / Max: 337.8 Min: 24.93 / Avg: 83.2 / Max: 132.66 Min: 80.15 / Avg: 185.44 / Max: 287.05
Zstd Compression This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Compression Speed EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.22, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.15, N = 3 SE +/- 0.09, N = 15 SE +/- 0.11, N = 15 SE +/- 0.13, N = 15 SE +/- 0.06, N = 3 SE +/- 0.12, N = 3 SE +/- 0.21, N = 4 SE +/- 0.03, N = 3 13.9 13.8 17.2 17.0 17.2 16.8 13.8 13.7 14.4 13.8 14.5 13.3 14.6 15.2 18.0 17.8 1. (CC) gcc options: -O3 -pthread -lz -llzma
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 300 600 900 1200 1500 SE +/- 2.09, N = 3 SE +/- 0.92, N = 3 SE +/- 1.17, N = 3 SE +/- 1.87, N = 3 SE +/- 1.99, N = 3 SE +/- 2.93, N = 3 SE +/- 1.95, N = 3 SE +/- 0.19, N = 3 SE +/- 0.56, N = 3 SE +/- 1.00, N = 3 SE +/- 1.45, N = 3 SE +/- 1.78, N = 15 SE +/- 0.43, N = 3 SE +/- 3.69, N = 3 SE +/- 0.84, N = 3 SE +/- 1.60, N = 3 1101.4 1096.5 1328.5 1324.6 1324.9 1321.4 1102.2 1104.2 1131.1 1041.8 1133.3 1046.0 1069.1 992.1 1242.6 1189.7 1. (CC) gcc options: -O3 -pthread -lz -llzma MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 20.514 20.797 13.268 5.643 11.661 5.089 14.042 5.816 5.787 2.512 5.674 2.539 6.931 2.956 9.241 4.173 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Zstd Compression 1.5.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 80 160 240 320 400 Min: 17.05 / Avg: 53.69 / Max: 60.18 Min: 15.43 / Avg: 52.72 / Max: 59.1 Min: 28.35 / Avg: 100.13 / Max: 109.97 Min: 57.79 / Avg: 234.74 / Max: 249.39 Min: 33.27 / Avg: 113.62 / Max: 124.55 Min: 64.16 / Avg: 259.66 / Max: 276.36 Min: 18.92 / Avg: 78.5 / Max: 87.36 Min: 39.89 / Avg: 189.86 / Max: 203.51 Min: 73.9 / Avg: 195.37 / Max: 210.14 Min: 208.83 / Avg: 414.8 / Max: 438.11 Min: 82.48 / Avg: 199.75 / Max: 214.36 Min: 201.8 / Avg: 412.05 / Max: 436.43 Min: 61.25 / Avg: 154.25 / Max: 170.09 Min: 172.05 / Avg: 335.67 / Max: 356.47 Min: 33.99 / Avg: 134.46 / Max: 153.43 Min: 75.89 / Avg: 285.1 / Max: 313.97
Timed Mesa Compilation This test profile times how long it takes to compile Mesa with Meson/Ninja. For minimizing build dependencies and avoid versioning conflicts, test this is just the core Mesa build without LLVM or the extra Gallium3D/Mesa drivers enabled. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Mesa Compilation 24.0 Time To Compile EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 4 SE +/- 0.02, N = 3 SE +/- 0.08, N = 4 SE +/- 0.03, N = 3 SE +/- 0.12, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.17, N = 3 SE +/- 0.07, N = 3 SE +/- 0.12, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 4 21.04 21.53 16.96 16.41 16.67 16.40 19.91 19.56 20.18 18.47 19.65 18.14 19.23 18.29 17.26 16.18 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Timed Mesa Compilation 24.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.48 / Avg: 79 / Max: 166.34 Min: 16.52 / Avg: 73.01 / Max: 143.27 Min: 27.72 / Avg: 132.61 / Max: 328.77 Min: 53.33 / Avg: 251.26 / Max: 671.83 Min: 33.71 / Avg: 153.28 / Max: 411.05 Min: 65.81 / Avg: 287.81 / Max: 802.37 Min: 20.4 / Avg: 109.13 / Max: 324.24 Min: 41.25 / Avg: 215.57 / Max: 638.97 Min: 74.89 / Avg: 232.74 / Max: 375.37 Min: 199.11 / Avg: 446.38 / Max: 751.99 Min: 73.22 / Avg: 228.31 / Max: 375.9 Min: 195.35 / Avg: 442.47 / Max: 753.73 Min: 57 / Avg: 172.35 / Max: 378.42 Min: 171.78 / Avg: 374.89 / Max: 753.6 Min: 32.86 / Avg: 176.64 / Max: 377.39 Min: 82.26 / Avg: 324.23 / Max: 753.39
Numpy Benchmark This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Score, More Is Better Numpy Benchmark EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 120 240 360 480 600 SE +/- 0.72, N = 3 SE +/- 1.34, N = 3 SE +/- 2.28, N = 3 SE +/- 0.38, N = 3 SE +/- 1.85, N = 3 SE +/- 0.60, N = 3 SE +/- 0.96, N = 3 SE +/- 0.25, N = 3 SE +/- 0.60, N = 3 SE +/- 1.13, N = 3 SE +/- 0.79, N = 3 SE +/- 1.30, N = 3 SE +/- 0.33, N = 3 SE +/- 0.26, N = 3 SE +/- 1.84, N = 3 SE +/- 1.42, N = 3 421.99 418.89 529.80 529.03 542.15 540.96 425.46 422.76 461.61 435.21 464.72 439.84 455.57 456.18 520.22 519.89 Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Numpy Benchmark EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 7.918 7.962 5.359 2.249 4.854 2.079 5.405 2.170 3.290 1.343 3.164 1.375 4.183 1.716 5.255 2.360 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Numpy Benchmark CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 80 160 240 320 400 Min: 16.02 / Avg: 53.3 / Max: 57.51 Min: 14.57 / Avg: 52.61 / Max: 56.75 Min: 26.33 / Avg: 98.86 / Max: 103.35 Min: 57.9 / Avg: 235.19 / Max: 242.43 Min: 32.28 / Avg: 111.7 / Max: 116.99 Min: 64.52 / Avg: 260.26 / Max: 267.94 Min: 20.13 / Avg: 78.72 / Max: 83.54 Min: 38.63 / Avg: 194.78 / Max: 202.23 Min: 73.46 / Avg: 140.29 / Max: 199.25 Min: 137.87 / Avg: 324.03 / Max: 425.17 Min: 82.95 / Avg: 146.89 / Max: 202.52 Min: 203.89 / Avg: 319.82 / Max: 421.94 Min: 59.12 / Avg: 108.92 / Max: 159.89 Min: 168.75 / Avg: 265.77 / Max: 348.44 Min: 33.91 / Avg: 98.99 / Max: 138.69 Min: 78.29 / Avg: 220.31 / Max: 293.25
libavif avifenc This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 0 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 20 40 60 80 100 SE +/- 0.19, N = 3 SE +/- 0.18, N = 3 SE +/- 0.01, N = 3 SE +/- 0.11, N = 3 SE +/- 0.12, N = 3 SE +/- 0.01, N = 3 SE +/- 0.13, N = 3 SE +/- 0.06, N = 3 SE +/- 0.15, N = 3 SE +/- 0.29, N = 3 SE +/- 0.22, N = 3 SE +/- 0.32, N = 3 SE +/- 0.32, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 75.86 78.24 61.99 62.89 61.63 63.40 73.27 74.69 76.92 69.65 73.06 68.44 69.93 67.38 64.48 60.53 1. (CXX) g++ options: -O3 -fPIC -lm CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better libavif avifenc 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.01 / Avg: 85.23 / Max: 156.74 Min: 16.02 / Avg: 77.61 / Max: 131.52 Min: 29.48 / Avg: 156.52 / Max: 318.36 Min: 58.14 / Avg: 295.3 / Max: 499.26 Min: 30.24 / Avg: 180.87 / Max: 389.03 Min: 62.98 / Avg: 324.3 / Max: 548.71 Min: 19.55 / Avg: 121.7 / Max: 248.44 Min: 39.64 / Avg: 234.99 / Max: 370.06 Min: 75.95 / Avg: 254.61 / Max: 372.81 Min: 205.63 / Avg: 494.12 / Max: 704.89 Min: 69.59 / Avg: 253.47 / Max: 376.81 Min: 208.88 / Avg: 485.4 / Max: 727.55 Min: 61.27 / Avg: 218.41 / Max: 375.95 Min: 152.89 / Avg: 425.04 / Max: 708.08 Min: 34.78 / Avg: 197.77 / Max: 373.2 Min: 77.6 / Avg: 391.09 / Max: 728.04
Result
OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 2 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 10 20 30 40 50 SE +/- 0.03, N = 3 SE +/- 0.10, N = 3 SE +/- 0.02, N = 3 SE +/- 0.43, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.13, N = 3 SE +/- 0.21, N = 3 SE +/- 0.22, N = 3 SE +/- 0.31, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.16, N = 3 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 41.66 42.61 34.15 35.40 34.04 35.32 40.35 41.92 41.13 38.65 39.88 38.14 37.75 37.31 34.79 33.65 1. (CXX) g++ options: -O3 -fPIC -lm CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better libavif avifenc 1.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 16.88 / Avg: 75.22 / Max: 155.11 Min: 14.11 / Avg: 69.45 / Max: 130.44 Min: 28.7 / Avg: 137.31 / Max: 316.42 Min: 57.84 / Avg: 267.46 / Max: 494.12 Min: 32.31 / Avg: 157.68 / Max: 383.9 Min: 63.92 / Avg: 296.47 / Max: 540.97 Min: 20.27 / Avg: 107.23 / Max: 247.8 Min: 39.83 / Avg: 217.6 / Max: 369.6 Min: 71.37 / Avg: 235.53 / Max: 369.9 Min: 207.34 / Avg: 461.85 / Max: 707.44 Min: 80.19 / Avg: 235.72 / Max: 375.98 Min: 195.24 / Avg: 458.52 / Max: 733.31 Min: 58.97 / Avg: 196.29 / Max: 374.92 Min: 149.52 / Avg: 390.93 / Max: 714.36 Min: 34.19 / Avg: 161.32 / Max: 375.73 Min: 77.42 / Avg: 350.42 / Max: 724.52
VVenC VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Fast EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 SE +/- 0.038, N = 3 SE +/- 0.041, N = 3 SE +/- 0.048, N = 3 SE +/- 0.057, N = 3 SE +/- 0.045, N = 3 SE +/- 0.010, N = 3 SE +/- 0.081, N = 3 SE +/- 0.033, N = 3 SE +/- 0.043, N = 15 SE +/- 0.022, N = 3 SE +/- 0.047, N = 3 SE +/- 0.023, N = 3 SE +/- 0.038, N = 3 6.118 6.000 7.431 7.547 6.465 6.324 6.023 6.501 6.077 6.719 6.480 7.392 6.960 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Fast EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0151 0.0302 0.0453 0.0604 0.0755 0.062 0.067 0.042 0.037 0.048 0.023 0.012 0.023 0.012 0.028 0.015 0.032 0.017 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better VVenC 1.11 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9684X EPYC 9754 Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 110 220 330 440 550 Min: 16.24 / Avg: 99.45 / Max: 145.04 Min: 15.3 / Avg: 89.82 / Max: 125.64 Min: 27.21 / Avg: 177.95 / Max: 263.84 Min: 31.5 / Avg: 204.88 / Max: 302.64 Min: 20.88 / Avg: 134.72 / Max: 195.13 Min: 71.41 / Avg: 273.3 / Max: 339.11 Min: 218.19 / Avg: 515.79 / Max: 625.23 Min: 84.83 / Avg: 277.48 / Max: 349.11 Min: 193.59 / Avg: 516.46 / Max: 637.28 Min: 59.59 / Avg: 242.83 / Max: 330.87 Min: 154.77 / Avg: 444.48 / Max: 576.57 Min: 26.5 / Avg: 232.82 / Max: 325.56 Min: 77.27 / Avg: 413.94 / Max: 550.57 Video Input: Bosphorus 4K - Video Preset: Fast
EPYC 9654 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1
EPYC 9684X 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1
EPYC 9754 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1
WebP Image Encode This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless, Highest Compression EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.135 0.27 0.405 0.54 0.675 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.48 0.48 0.57 0.57 0.57 0.56 0.48 0.48 0.55 0.54 0.55 0.54 0.52 0.51 0.60 0.59 1. (CC) gcc options: -fvisibility=hidden -O2 -lm MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless, Highest Compression EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.002 0.004 0.006 0.008 0.01 0.009 0.009 0.006 0.003 0.005 0.002 0.006 0.003 0.003 0.001 0.003 0.001 0.004 0.002 0.005 0.002 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better WebP Image Encode 1.2.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 80 160 240 320 400 Min: 15.79 / Avg: 53.42 / Max: 58.48 Min: 15.76 / Avg: 52.31 / Max: 56.63 Min: 27.22 / Avg: 96.75 / Max: 105.57 Min: 57.93 / Avg: 225.88 / Max: 245.29 Min: 30.35 / Avg: 109.81 / Max: 119.47 Min: 61.65 / Avg: 249.88 / Max: 268.66 Min: 20.24 / Avg: 77.06 / Max: 83.97 Min: 39.58 / Avg: 186.91 / Max: 199.67 Min: 80.95 / Avg: 184.91 / Max: 202.94 Min: 205.48 / Avg: 407.7 / Max: 432.16 Min: 83.4 / Avg: 190.72 / Max: 207.25 Min: 205.98 / Avg: 397.44 / Max: 422.1 Min: 60.77 / Avg: 148.44 / Max: 161.11 Min: 150.62 / Avg: 324.76 / Max: 344.8 Min: 34.5 / Avg: 128.71 / Max: 144.36 Min: 79.69 / Avg: 274 / Max: 299.87
SecureMark SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions developed by EEMBC. SecureMark-TLS is benchmarking Transport Layer Security performance with a focus on IoT/edge computing. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org marks, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70K 140K 210K 280K 350K SE +/- 372.96, N = 3 SE +/- 542.88, N = 3 SE +/- 361.20, N = 3 SE +/- 389.81, N = 3 SE +/- 415.71, N = 3 SE +/- 809.90, N = 3 SE +/- 457.42, N = 3 SE +/- 553.62, N = 3 SE +/- 256.85, N = 3 SE +/- 384.08, N = 3 SE +/- 670.43, N = 3 SE +/- 196.24, N = 3 SE +/- 625.56, N = 3 SE +/- 330.95, N = 3 SE +/- 209.97, N = 3 SE +/- 411.11, N = 3 250024 250162 298341 299610 300833 301740 250695 251075 277647 277589 277260 277298 277205 275472 309896 308434 1. (CC) gcc options: -pedantic -O3 marks Per Watt
OpenBenchmarking.org marks Per Watt, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 1000 2000 3000 4000 5000 4697.25 4781.86 3034.03 1272.88 2686.16 1152.38 3204.23 1278.59 2308.76 986.37 2198.71 973.55 3029.20 1185.32 3643.38 1242.56 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better SecureMark 1.0.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 60 120 180 240 300 Min: 16.16 / Avg: 53.23 / Max: 56.82 Min: 14.73 / Avg: 52.31 / Max: 55.82 Min: 28.33 / Avg: 98.33 / Max: 101.2 Min: 56.74 / Avg: 235.38 / Max: 240.55 Min: 31.06 / Avg: 111.99 / Max: 116.51 Min: 60.77 / Avg: 261.84 / Max: 267.2 Min: 18.81 / Avg: 78.24 / Max: 81.48 Min: 41.92 / Avg: 196.37 / Max: 200.73 Min: 74.22 / Avg: 120.26 / Max: 126.65 Min: 137.33 / Avg: 281.43 / Max: 342.59 Min: 85.74 / Avg: 126.1 / Max: 134.53 Min: 124.02 / Avg: 284.83 / Max: 338.05 Min: 53.54 / Avg: 91.51 / Max: 104.77 Min: 171.99 / Avg: 232.4 / Max: 318.93 Min: 33.51 / Avg: 85.06 / Max: 96.45 Min: 78.38 / Avg: 248.23 / Max: 290.1
Google SynthMark SynthMark is a cross platform tool for benchmarking CPU performance under a variety of real-time audio workloads. It uses a polyphonic synthesizer model to provide standardized tests for latency, jitter and computational throughput. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Voices, More Is Better Google SynthMark 20201109 Test: VoiceMark_100 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 200 400 600 800 1000 SE +/- 0.14, N = 3 SE +/- 0.60, N = 3 SE +/- 0.55, N = 3 SE +/- 0.56, N = 3 SE +/- 0.69, N = 3 SE +/- 0.72, N = 3 SE +/- 0.45, N = 3 SE +/- 0.88, N = 3 SE +/- 0.26, N = 3 SE +/- 0.30, N = 3 SE +/- 0.24, N = 3 SE +/- 0.48, N = 3 SE +/- 0.22, N = 3 SE +/- 1.51, N = 3 SE +/- 0.13, N = 3 SE +/- 0.38, N = 3 655.24 656.25 783.78 782.32 783.42 783.20 655.80 655.05 711.10 710.92 712.01 712.09 711.38 712.70 799.72 799.49 1. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast Voices Per Watt
OpenBenchmarking.org Voices Per Watt, More Is Better Google SynthMark 20201109 Test: VoiceMark_100 EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 12.984 13.295 8.699 3.607 7.582 3.246 9.006 3.610 6.250 2.322 5.891 2.567 7.626 3.044 7.777 3.121 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Google SynthMark 20201109 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 Min: 17.96 / Avg: 50.47 / Max: 55.88 Min: 16.18 / Avg: 49.36 / Max: 54.66 Min: 28.69 / Avg: 90.1 / Max: 99.51 Min: 57.99 / Avg: 216.87 / Max: 238.45 Min: 33.97 / Avg: 103.33 / Max: 113.89 Min: 61.62 / Avg: 241.28 / Max: 266.15 Min: 21.02 / Avg: 72.82 / Max: 80.06 Min: 40.35 / Avg: 181.43 / Max: 200.97 Min: 75.06 / Avg: 113.78 / Max: 122.61 Min: 203.32 / Avg: 306.14 / Max: 413.85 Min: 84.65 / Avg: 120.87 / Max: 129.59 Min: 201.92 / Avg: 277.45 / Max: 348.85 Min: 63.55 / Avg: 93.65 / Max: 130.75 Min: 172.36 / Avg: 234.1 / Max: 329.35 Min: 33.54 / Avg: 102.83 / Max: 128.91 Min: 77.31 / Avg: 256.19 / Max: 292.81
WebP Image Encode This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Highest Compression EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.8145 1.629 2.4435 3.258 4.0725 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 3.03 3.03 3.62 3.62 3.61 3.62 3.04 3.03 3.10 3.09 3.10 3.09 3.08 3.07 3.47 3.46 1. (CC) gcc options: -fvisibility=hidden -O2 -lm MP/s Per Watt
OpenBenchmarking.org MP/s Per Watt, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Highest Compression EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0162 0.0324 0.0486 0.0648 0.081 0.069 0.072 0.048 0.021 0.044 0.019 0.049 0.020 0.028 0.011 0.026 0.011 0.035 0.014 0.051 0.019 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better WebP Image Encode 1.2.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 Min: 16.45 / Avg: 44.16 / Max: 57.18 Min: 16.53 / Avg: 42.26 / Max: 55.05 Min: 28.98 / Avg: 75.95 / Max: 101.21 Min: 56.66 / Avg: 176 / Max: 238.56 Min: 32.65 / Avg: 82.63 / Max: 115.67 Min: 63.07 / Avg: 192.89 / Max: 262.1 Min: 20.53 / Avg: 61.84 / Max: 81.06 Min: 39.42 / Avg: 149.38 / Max: 197.42 Min: 70.24 / Avg: 111.05 / Max: 144.04 Min: 200.76 / Avg: 274.15 / Max: 348.74 Min: 81.36 / Avg: 119.53 / Max: 163.62 Min: 205.94 / Avg: 277 / Max: 371.07 Min: 59.48 / Avg: 88.57 / Max: 151.52 Min: 148.59 / Avg: 222.03 / Max: 316.69 Min: 34.25 / Avg: 68.3 / Max: 101.37 Min: 74.68 / Avg: 186.13 / Max: 288.17
srsRAN Project srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mbps, More Is Better srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Thread EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 112.1 112.1 112.2 112.2 112.2 112.2 112.1 112.1 129.7 129.7 129.7 129.7 129.7 129.7 129.5 129.5 MIN: 79.3 MIN: 79.3 MIN: 79.3 MIN: 79.3 MIN: 79.3 MIN: 79.3 MIN: 79.3 MIN: 79.3 MIN: 105.9 MIN: 105.9 MIN: 105.9 MIN: 105.9 MIN: 105.9 MIN: 105.9 MIN: 105.7 MIN: 105.7 1. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl Mbps Per Watt
OpenBenchmarking.org Mbps Per Watt, More Is Better srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Thread EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.511 1.022 1.533 2.044 2.555 2.234 2.271 1.257 0.524 1.095 0.476 1.551 0.637 0.750 0.345 0.736 0.347 0.965 0.426 1.196 0.564 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 70 140 210 280 350 Min: 16.97 / Avg: 50.19 / Max: 57.02 Min: 16.95 / Avg: 49.36 / Max: 56.04 Min: 27.03 / Avg: 89.27 / Max: 101.62 Min: 58.78 / Avg: 213.93 / Max: 242.48 Min: 34.3 / Avg: 102.46 / Max: 115.98 Min: 61.91 / Avg: 235.89 / Max: 267.82 Min: 21.9 / Avg: 72.27 / Max: 82.25 Min: 40.92 / Avg: 175.84 / Max: 200.95 Min: 72.46 / Avg: 172.93 / Max: 196.41 Min: 207.62 / Avg: 375.78 / Max: 415.17 Min: 77.25 / Avg: 176.34 / Max: 199.62 Min: 206.09 / Avg: 373.48 / Max: 412.46 Min: 59.93 / Avg: 134.44 / Max: 153.53 Min: 160.99 / Avg: 304.32 / Max: 339.38 Min: 34.52 / Avg: 108.29 / Max: 130.82 Min: 80.8 / Avg: 229.54 / Max: 284.84
CPU Power Consumption Monitor OpenBenchmarking.org Watts CPU Power Consumption Monitor Phoronix Test Suite System Monitoring EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 170 340 510 680 850 Min: 8.43 / Avg: 128.7 / Max: 184.62 Min: 9.26 / Avg: 114.13 / Max: 155.71 Min: 14.69 / Avg: 244.69 / Max: 373.1 Min: 35.7 / Avg: 464.49 / Max: 751.79 Min: 17.06 / Avg: 290.08 / Max: 413.51 Min: 33.66 / Avg: 525.49 / Max: 837.69 Min: 10.09 / Avg: 225.05 / Max: 397.44 Min: 20.22 / Avg: 375.51 / Max: 735.74 Min: 38.11 / Avg: 291.53 / Max: 379.76 Min: 102.49 / Avg: 566.6 / Max: 756.48 Min: 38.16 / Avg: 295.53 / Max: 384.53 Min: 95.52 / Avg: 569.8 / Max: 760.61 Min: 30.38 / Avg: 278.1 / Max: 383.24 Min: 87.48 / Avg: 517.33 / Max: 758.17 Min: 17.77 / Avg: 260.08 / Max: 405.42 Min: 39.72 / Avg: 455.07 / Max: 946.96
Stockfish OpenBenchmarking.org Watts, Fewer Is Better Stockfish 16.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.26 / Avg: 174.36 / Max: 184.62 Min: 16.19 / Avg: 145.98 / Max: 155.71 Min: 30.05 / Avg: 347.3 / Max: 373.1 Min: 57.4 / Avg: 686.5 / Max: 751.79 Min: 35.5 / Avg: 386.38 / Max: 410.66 Min: 70.77 / Avg: 766.5 / Max: 822.06 Min: 21.51 / Avg: 332.06 / Max: 363.69 Min: 37.25 / Avg: 640.56 / Max: 735.74 Min: 71.31 / Avg: 341.39 / Max: 379.06 Min: 203.83 / Avg: 679.75 / Max: 756.48 Min: 74.29 / Avg: 341.12 / Max: 366.15 Min: 195.85 / Avg: 680.45 / Max: 758.53 Min: 56.83 / Avg: 339.56 / Max: 379.91 Min: 166.29 / Avg: 678.39 / Max: 758.17 Min: 22.35 / Avg: 337.72 / Max: 370.84 Min: 78.55 / Avg: 672.28 / Max: 755.5
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better Stockfish 16.1 Chess Benchmark EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130K 260K 390K 520K 650K 568916.36 621037.11 569277.48 540060.41 555213.84 520707.98 571475.98 470216.10 199243.06 202981.45 215737.78 219573.26 285328.34 276776.39 325017.74 331127.60
Result
OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 16.1 Chess Benchmark EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90M 180M 270M 360M 450M SE +/- 2397455.91, N = 12 SE +/- 2147688.02, N = 12 SE +/- 3170744.86, N = 15 SE +/- 2356843.91, N = 3 SE +/- 2268830.04, N = 3 SE +/- 919073.22, N = 3 SE +/- 4652276.19, N = 12 SE +/- 7961094.38, N = 12 SE +/- 1751369.24, N = 12 SE +/- 4344961.90, N = 10 SE +/- 629927.17, N = 3 SE +/- 2925715.37, N = 11 SE +/- 1304590.97, N = 3 SE +/- 2148606.38, N = 4 SE +/- 2966712.90, N = 9 SE +/- 2482737.87, N = 3 99198144 90658156 197708350 370751530 214524659 399124223 189763799 301202232 68019088 137977203 73592230 149409082 96887479 187761769 109766345 222610378 1. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -msse -msse3 -mpopcnt -mavx2 -mbmi -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto-partition=one -flto=jobserver Nodes Per Second Per Watt
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better Stockfish 16.1 Chess Benchmark EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130K 260K 390K 520K 650K 568916.36 621037.11 569277.48 540060.41 555213.84 520707.98 571475.98 470216.10 199243.06 202981.45 215737.78 219573.26 285328.34 276776.39 325017.74 331127.60 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Stockfish 16.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.26 / Avg: 174.36 / Max: 184.62 Min: 16.19 / Avg: 145.98 / Max: 155.71 Min: 30.05 / Avg: 347.3 / Max: 373.1 Min: 57.4 / Avg: 686.5 / Max: 751.79 Min: 35.5 / Avg: 386.38 / Max: 410.66 Min: 70.77 / Avg: 766.5 / Max: 822.06 Min: 21.51 / Avg: 332.06 / Max: 363.69 Min: 37.25 / Avg: 640.56 / Max: 735.74 Min: 71.31 / Avg: 341.39 / Max: 379.06 Min: 203.83 / Avg: 679.75 / Max: 756.48 Min: 74.29 / Avg: 341.12 / Max: 366.15 Min: 195.85 / Avg: 680.45 / Max: 758.53 Min: 56.83 / Avg: 339.56 / Max: 379.91 Min: 166.29 / Avg: 678.39 / Max: 758.17 Min: 22.35 / Avg: 337.72 / Max: 370.84 Min: 78.55 / Avg: 672.28 / Max: 755.5
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8 16 24 32 40 SE +/- 0.63, N = 15 SE +/- 0.61, N = 15 SE +/- 0.50, N = 15 SE +/- 0.61, N = 15 SE +/- 0.61, N = 15 SE +/- 0.40, N = 15 SE +/- 0.45, N = 15 SE +/- 0.23, N = 15 SE +/- 0.11, N = 3 SE +/- 0.40, N = 3 SE +/- 0.31, N = 3 SE +/- 0.33, N = 3 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 27.93 26.83 23.92 25.32 21.58 22.79 26.45 27.90 33.33 36.23 33.99 36.00 29.41 29.75 29.02 28.85 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.17 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 11 22 33 44 55 SE +/- 0.87, N = 15 SE +/- 0.82, N = 15 SE +/- 0.91, N = 15 SE +/- 0.92, N = 15 SE +/- 1.36, N = 15 SE +/- 0.76, N = 15 SE +/- 0.63, N = 15 SE +/- 0.27, N = 15 SE +/- 0.09, N = 3 SE +/- 0.31, N = 3 SE +/- 0.27, N = 3 SE +/- 0.26, N = 3 SE +/- 0.05, N = 3 SE +/- 0.13, N = 3 SE +/- 0.10, N = 3 SE +/- 0.11, N = 3 36.08 37.53 42.07 39.80 46.88 44.06 37.96 35.87 30.00 27.61 29.43 27.78 34.00 33.61 34.46 34.66 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 8 16 24 32 40 SE +/- 0.63, N = 15 SE +/- 0.61, N = 15 SE +/- 0.50, N = 15 SE +/- 0.61, N = 15 SE +/- 0.61, N = 15 SE +/- 0.40, N = 15 SE +/- 0.45, N = 15 SE +/- 0.23, N = 15 SE +/- 0.11, N = 3 SE +/- 0.40, N = 3 SE +/- 0.31, N = 3 SE +/- 0.33, N = 3 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 27.93 26.83 23.92 25.32 21.58 22.79 26.45 27.90 33.33 36.23 33.99 36.00 29.41 29.75 29.02 28.85 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 6.87, N = 12 SE +/- 7.19, N = 15 SE +/- 4.84, N = 15 SE +/- 0.35, N = 3 SE +/- 4.76, N = 12 SE +/- 2.18, N = 15 SE +/- 2.29, N = 15 SE +/- 2.91, N = 3 SE +/- 3.36, N = 15 SE +/- 5.82, N = 12 SE +/- 6.41, N = 15 SE +/- 2.59, N = 15 SE +/- 0.09, N = 3 SE +/- 0.69, N = 15 SE +/- 0.71, N = 3 SE +/- 0.44, N = 3 216.70 226.63 182.02 220.35 178.03 182.27 166.79 203.57 139.10 121.67 155.79 121.78 116.32 92.35 109.67 100.74 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Result
OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.17 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 3 6 9 12 15 SE +/- 0.14467, N = 12 SE +/- 0.11914, N = 15 SE +/- 0.15150, N = 15 SE +/- 0.00718, N = 3 SE +/- 0.16422, N = 12 SE +/- 0.06453, N = 15 SE +/- 0.08489, N = 15 SE +/- 0.07023, N = 3 SE +/- 0.15651, N = 15 SE +/- 0.34632, N = 12 SE +/- 0.22869, N = 15 SE +/- 0.16487, N = 15 SE +/- 0.00664, N = 3 SE +/- 0.07619, N = 15 SE +/- 0.05824, N = 3 SE +/- 0.04271, N = 3 4.66491 4.46533 5.55018 4.53817 5.66508 5.49711 6.01164 4.91420 7.24187 8.39962 6.54965 8.26016 8.59719 10.83623 9.11885 9.92684 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt Inference Time Cost (ms)
OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.17 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 50 100 150 200 250 SE +/- 6.87, N = 12 SE +/- 7.19, N = 15 SE +/- 4.84, N = 15 SE +/- 0.35, N = 3 SE +/- 4.76, N = 12 SE +/- 2.18, N = 15 SE +/- 2.29, N = 15 SE +/- 2.91, N = 3 SE +/- 3.36, N = 15 SE +/- 5.82, N = 12 SE +/- 6.41, N = 15 SE +/- 2.59, N = 15 SE +/- 0.09, N = 3 SE +/- 0.69, N = 15 SE +/- 0.71, N = 3 SE +/- 0.44, N = 3 216.70 226.63 182.02 220.35 178.03 182.27 166.79 203.57 139.10 121.67 155.79 121.78 116.32 92.35 109.67 100.74 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
srsRAN Project OpenBenchmarking.org Watts, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.85 / Avg: 136.58 / Max: 166.81 Min: 16.71 / Avg: 117.93 / Max: 142.26 Min: 28.12 / Avg: 276.24 / Max: 341.73 Min: 55.25 / Avg: 559.64 / Max: 633.66 Min: 32.26 / Avg: 321.41 / Max: 408.75 Min: 65.52 / Avg: 632.5 / Max: 723.62 Min: 21.57 / Avg: 261.04 / Max: 306.84 Min: 41.23 / Avg: 514.74 / Max: 612.62 Min: 74.5 / Avg: 281.29 / Max: 379.76 Min: 195.95 / Avg: 604.92 / Max: 745.23 Min: 74.87 / Avg: 289.12 / Max: 373.87 Min: 189.97 / Avg: 617.8 / Max: 721.97 Min: 60.24 / Avg: 283.1 / Max: 369.61 Min: 148.45 / Avg: 589.02 / Max: 721.35 Min: 25.9 / Avg: 266.9 / Max: 379.06 Min: 68.27 / Avg: 537.03 / Max: 710
OpenBenchmarking.org Mbps Per Watt, More Is Better srsRAN Project 23.10.1-20240325 Test: PDSCH Processor Benchmark, Throughput Total EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 141.87 165.23 100.26 40.98 91.42 41.07 96.52 58.04 75.68 49.98 52.16 19.21 87.21 51.12 98.83 56.88
Result
OpenBenchmarking.org Mbps, More Is Better srsRAN Project 23.10.1-20240325 Test: PDSCH Processor Benchmark, Throughput Total EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 7K 14K 21K 28K 35K SE +/- 207.19, N = 5 SE +/- 58.21, N = 3 SE +/- 357.13, N = 12 SE +/- 526.01, N = 15 SE +/- 348.28, N = 4 SE +/- 343.97, N = 3 SE +/- 237.63, N = 3 SE +/- 2115.57, N = 15 SE +/- 938.95, N = 15 SE +/- 175.78, N = 4 SE +/- 281.72, N = 15 SE +/- 490.54, N = 15 SE +/- 236.27, N = 4 SE +/- 312.17, N = 3 SE +/- 298.34, N = 15 SE +/- 356.81, N = 15 19375.9 19485.8 27695.2 22931.8 29384.9 25977.4 25194.6 29877.1 21286.6 30236.0 15081.0 11866.7 24688.5 30108.7 26378.1 30547.6 1. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl Mbps Per Watt
OpenBenchmarking.org Mbps Per Watt, More Is Better srsRAN Project 23.10.1-20240325 Test: PDSCH Processor Benchmark, Throughput Total EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 40 80 120 160 200 141.87 165.23 100.26 40.98 91.42 41.07 96.52 58.04 75.68 49.98 52.16 19.21 87.21 51.12 98.83 56.88 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 17.85 / Avg: 136.58 / Max: 166.81 Min: 16.71 / Avg: 117.93 / Max: 142.26 Min: 28.12 / Avg: 276.24 / Max: 341.73 Min: 55.25 / Avg: 559.64 / Max: 633.66 Min: 32.26 / Avg: 321.41 / Max: 408.75 Min: 65.52 / Avg: 632.5 / Max: 723.62 Min: 21.57 / Avg: 261.04 / Max: 306.84 Min: 41.23 / Avg: 514.74 / Max: 612.62 Min: 74.5 / Avg: 281.29 / Max: 379.76 Min: 195.95 / Avg: 604.92 / Max: 745.23 Min: 74.87 / Avg: 289.12 / Max: 373.87 Min: 189.97 / Avg: 617.8 / Max: 721.97 Min: 60.24 / Avg: 283.1 / Max: 369.61 Min: 148.45 / Avg: 589.02 / Max: 721.35 Min: 25.9 / Avg: 266.9 / Max: 379.06 Min: 68.27 / Avg: 537.03 / Max: 710
LuxCoreRender OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 150 300 450 600 750 Min: 18.62 / Avg: 156.14 / Max: 171.55 Min: 18.19 / Avg: 128.95 / Max: 140.58 Min: 29.9 / Avg: 299.39 / Max: 326.71 Min: 59.66 / Avg: 573.93 / Max: 661.76 Min: 34.75 / Avg: 384.12 / Max: 412.45 Min: 65.89 / Avg: 739.45 / Max: 837.69 Min: 20.73 / Avg: 311.89 / Max: 340.37 Min: 42.95 / Avg: 561.1 / Max: 675.69 Min: 74.56 / Avg: 331.26 / Max: 374.04 Min: 195.7 / Avg: 665.76 / Max: 753.2 Min: 72.86 / Avg: 330.29 / Max: 377.65 Min: 203.55 / Avg: 668.4 / Max: 754.98 Min: 59.67 / Avg: 330.53 / Max: 373.83 Min: 173.98 / Avg: 664.56 / Max: 753.01 Min: 24.28 / Avg: 326.43 / Max: 375.69 Min: 76.68 / Avg: 654.92 / Max: 755.35
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0149 0.0298 0.0447 0.0596 0.0745 0.060 0.066 0.048 0.031 0.042 0.031 0.054 0.039 0.023 0.015 0.024 0.016 0.026 0.017 0.029 0.019
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.06, N = 15 SE +/- 0.03, N = 3 SE +/- 0.13, N = 3 SE +/- 0.12, N = 3 SE +/- 0.18, N = 15 SE +/- 0.91, N = 15 SE +/- 0.28, N = 15 SE +/- 1.02, N = 15 SE +/- 0.08, N = 3 SE +/- 0.12, N = 3 SE +/- 0.10, N = 3 SE +/- 0.20, N = 15 SE +/- 0.06, N = 15 SE +/- 0.23, N = 15 SE +/- 0.02, N = 3 SE +/- 0.24, N = 15 9.39 8.48 14.29 17.89 16.21 22.58 16.79 22.00 7.46 10.17 8.00 10.44 8.72 11.12 9.38 12.74 MIN: 8.9 / MAX: 10.42 MIN: 8.26 / MAX: 9.41 MIN: 13.78 / MAX: 17.04 MIN: 16.29 / MAX: 28.22 MIN: 15.2 / MAX: 19.13 MIN: 17.46 / MAX: 32.78 MIN: 15.23 / MAX: 21.36 MIN: 16.32 / MAX: 35.07 MIN: 7.17 / MAX: 8.21 MIN: 9.34 / MAX: 14.75 MIN: 7.65 / MAX: 9.12 MIN: 8.86 / MAX: 16.08 MIN: 8.12 / MAX: 10.6 MIN: 9.18 / MAX: 17.51 MIN: 8.94 / MAX: 11.49 MIN: 10.69 / MAX: 19.19 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0149 0.0298 0.0447 0.0596 0.0745 0.060 0.066 0.048 0.031 0.042 0.031 0.054 0.039 0.023 0.015 0.024 0.016 0.026 0.017 0.029 0.019 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 150 300 450 600 750 Min: 18.62 / Avg: 156.14 / Max: 171.55 Min: 18.19 / Avg: 128.95 / Max: 140.58 Min: 29.9 / Avg: 299.39 / Max: 326.71 Min: 59.66 / Avg: 573.93 / Max: 661.76 Min: 34.75 / Avg: 384.12 / Max: 412.45 Min: 65.89 / Avg: 739.45 / Max: 837.69 Min: 20.73 / Avg: 311.89 / Max: 340.37 Min: 42.95 / Avg: 561.1 / Max: 675.69 Min: 74.56 / Avg: 331.26 / Max: 374.04 Min: 195.7 / Avg: 665.76 / Max: 753.2 Min: 72.86 / Avg: 330.29 / Max: 377.65 Min: 203.55 / Avg: 668.4 / Max: 754.98 Min: 59.67 / Avg: 330.53 / Max: 373.83 Min: 173.98 / Avg: 664.56 / Max: 753.01 Min: 24.28 / Avg: 326.43 / Max: 375.69 Min: 76.68 / Avg: 654.92 / Max: 755.35
NAS Parallel Benchmarks OpenBenchmarking.org Watts, Fewer Is Better NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.8 / Avg: 129.08 / Max: 158.73 Min: 17.1 / Avg: 113.27 / Max: 135.62 Min: 26.47 / Avg: 242.89 / Max: 334.43 Min: 56.44 / Avg: 416.97 / Max: 683.16 Min: 31.93 / Avg: 307.77 / Max: 409.54 Min: 65.8 / Avg: 495.73 / Max: 819.01 Min: 21.19 / Avg: 234.01 / Max: 336.9 Min: 40.5 / Avg: 379.03 / Max: 677.9 Min: 73.35 / Avg: 254.67 / Max: 351.04 Min: 200.97 / Avg: 537.97 / Max: 722.6 Min: 76.21 / Avg: 270.13 / Max: 345.76 Min: 196.31 / Avg: 561.84 / Max: 754.16 Min: 57.58 / Avg: 291.31 / Max: 376.98 Min: 155.56 / Avg: 520.43 / Max: 752.57 Min: 26.35 / Avg: 276.68 / Max: 375.06 Min: 68.69 / Avg: 471.7 / Max: 753.27
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 13 26 39 52 65 49.71 50.72 43.86 46.56 37.12 44.72 53.74 58.09 21.07 20.39 23.86 22.77 25.96 28.72 29.24 33.93
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5K 10K 15K 20K 25K SE +/- 14.68, N = 3 SE +/- 50.01, N = 8 SE +/- 122.47, N = 15 SE +/- 530.31, N = 15 SE +/- 309.03, N = 15 SE +/- 602.37, N = 15 SE +/- 367.88, N = 12 SE +/- 1274.85, N = 12 SE +/- 155.10, N = 12 SE +/- 4.83, N = 4 SE +/- 6.53, N = 3 SE +/- 10.93, N = 4 SE +/- 6.77, N = 3 SE +/- 100.23, N = 4 SE +/- 22.33, N = 3 SE +/- 30.19, N = 4 6416.34 5745.04 10652.03 19414.94 11424.05 22166.60 12575.30 22017.75 5366.69 10970.45 6445.09 12793.01 7562.80 14946.76 8089.14 16004.07 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6 Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 13 26 39 52 65 49.71 50.72 43.86 46.56 37.12 44.72 53.74 58.09 21.07 20.39 23.86 22.77 25.96 28.72 29.24 33.93 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.8 / Avg: 129.08 / Max: 158.73 Min: 17.1 / Avg: 113.27 / Max: 135.62 Min: 26.47 / Avg: 242.89 / Max: 334.43 Min: 56.44 / Avg: 416.97 / Max: 683.16 Min: 31.93 / Avg: 307.77 / Max: 409.54 Min: 65.8 / Avg: 495.73 / Max: 819.01 Min: 21.19 / Avg: 234.01 / Max: 336.9 Min: 40.5 / Avg: 379.03 / Max: 677.9 Min: 73.35 / Avg: 254.67 / Max: 351.04 Min: 200.97 / Avg: 537.97 / Max: 722.6 Min: 76.21 / Avg: 270.13 / Max: 345.76 Min: 196.31 / Avg: 561.84 / Max: 754.16 Min: 57.58 / Avg: 291.31 / Max: 376.98 Min: 155.56 / Avg: 520.43 / Max: 752.57 Min: 26.35 / Avg: 276.68 / Max: 375.06 Min: 68.69 / Avg: 471.7 / Max: 753.27
Pennant OpenBenchmarking.org Watts, Fewer Is Better Pennant 1.0.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.14 / Avg: 107.18 / Max: 164.25 Min: 16.11 / Avg: 95.67 / Max: 145.23 Min: 27.07 / Avg: 170.76 / Max: 330.73 Min: 54.46 / Avg: 322.18 / Max: 670.5 Min: 33.15 / Avg: 196.35 / Max: 409.53 Min: 63.33 / Avg: 370.49 / Max: 818.97 Min: 20.01 / Avg: 156.67 / Max: 328.66 Min: 41.42 / Avg: 291.74 / Max: 665.7 Min: 71.8 / Avg: 241.62 / Max: 375.61 Min: 187.85 / Avg: 448.22 / Max: 752.01 Min: 74.45 / Avg: 234.93 / Max: 375.6 Min: 191.18 / Avg: 450.17 / Max: 753.29 Min: 57.31 / Avg: 222.1 / Max: 377.34 Min: 144.69 / Avg: 414.55 / Max: 753.5 Min: 25.17 / Avg: 191.11 / Max: 377.76 Min: 64.47 / Avg: 342.65 / Max: 753.04
Result
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: leblancbig EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 SE +/- 0.057580, N = 6 SE +/- 0.045713, N = 13 SE +/- 0.115723, N = 15 SE +/- 0.063684, N = 15 SE +/- 0.115724, N = 15 SE +/- 0.055800, N = 15 SE +/- 0.115702, N = 15 SE +/- 0.022246, N = 15 SE +/- 0.025580, N = 6 SE +/- 0.010204, N = 7 SE +/- 0.023904, N = 7 SE +/- 0.008354, N = 7 SE +/- 0.022393, N = 7 SE +/- 0.019069, N = 15 SE +/- 0.022698, N = 7 SE +/- 0.013568, N = 6 6.541768 6.852712 3.754835 2.312977 3.360453 2.076835 3.654303 2.084384 5.930957 2.881099 5.114161 2.527363 4.955865 2.529807 4.182707 2.159459 1. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better Pennant 1.0.1 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.14 / Avg: 107.18 / Max: 164.25 Min: 16.11 / Avg: 95.67 / Max: 145.23 Min: 27.07 / Avg: 170.76 / Max: 330.73 Min: 54.46 / Avg: 322.18 / Max: 670.5 Min: 33.15 / Avg: 196.35 / Max: 409.53 Min: 63.33 / Avg: 370.49 / Max: 818.97 Min: 20.01 / Avg: 156.67 / Max: 328.66 Min: 41.42 / Avg: 291.74 / Max: 665.7 Min: 71.8 / Avg: 241.62 / Max: 375.61 Min: 187.85 / Avg: 448.22 / Max: 752.01 Min: 74.45 / Avg: 234.93 / Max: 375.6 Min: 191.18 / Avg: 450.17 / Max: 753.29 Min: 57.31 / Avg: 222.1 / Max: 377.34 Min: 144.69 / Avg: 414.55 / Max: 753.5 Min: 25.17 / Avg: 191.11 / Max: 377.76 Min: 64.47 / Avg: 342.65 / Max: 753.04
NAMD OpenBenchmarking.org Watts, Fewer Is Better NAMD 3.0b6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.33 / Avg: 130.18 / Max: 164.25 Min: 15.81 / Avg: 114.62 / Max: 140.3 Min: 28.55 / Avg: 235.77 / Max: 330 Min: 57.51 / Avg: 289.28 / Max: 672.46 Min: 32.28 / Avg: 274.76 / Max: 406.72 Min: 65.49 / Avg: 343.59 / Max: 811.63 Min: 21.48 / Avg: 206.2 / Max: 309.47 Min: 42.36 / Avg: 249.94 / Max: 651.86 Min: 71.79 / Avg: 281.33 / Max: 364.12 Min: 192.28 / Avg: 560.13 / Max: 714.79 Min: 72.24 / Avg: 295.54 / Max: 371.41 Min: 194.25 / Avg: 565.07 / Max: 686.05 Min: 55.23 / Avg: 273.91 / Max: 344.33 Min: 163.14 / Avg: 552.42 / Max: 653.46 Min: 33.09 / Avg: 260.62 / Max: 349.3 Min: 73.13 / Avg: 510.74 / Max: 686.96
OpenBenchmarking.org ns/day Per Watt, More Is Better NAMD 3.0b6 Input: STMV with 1,066,628 Atoms EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0052 0.0104 0.0156 0.0208 0.026 0.012 0.013 0.012 0.018 0.012 0.019 0.014 0.023 0.005 0.003 0.005 0.003 0.004 0.002 0.006 0.004
Result
OpenBenchmarking.org ns/day, More Is Better NAMD 3.0b6 Input: STMV with 1,066,628 Atoms EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 2 4 6 8 10 SE +/- 0.00440, N = 3 SE +/- 0.00540, N = 3 SE +/- 0.00537, N = 3 SE +/- 0.02927, N = 3 SE +/- 0.00206, N = 3 SE +/- 0.01739, N = 3 SE +/- 0.00847, N = 3 SE +/- 0.00332, N = 3 SE +/- 0.05087, N = 15 SE +/- 0.05644, N = 12 SE +/- 0.05862, N = 15 SE +/- 0.05463, N = 15 SE +/- 0.01396, N = 15 SE +/- 0.04403, N = 15 SE +/- 0.02648, N = 15 SE +/- 0.08185, N = 13 1.51901 1.46163 2.87877 5.27839 3.41085 6.38343 2.88869 5.83606 1.45223 1.88679 1.50214 1.76778 1.05274 1.10900 1.43787 2.25422 ns/day Per Watt
OpenBenchmarking.org ns/day Per Watt, More Is Better NAMD 3.0b6 Input: STMV with 1,066,628 Atoms EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0052 0.0104 0.0156 0.0208 0.026 0.012 0.013 0.012 0.018 0.012 0.019 0.014 0.023 0.005 0.003 0.005 0.003 0.004 0.002 0.006 0.004 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better NAMD 3.0b6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 18.33 / Avg: 130.18 / Max: 164.25 Min: 15.81 / Avg: 114.62 / Max: 140.3 Min: 28.55 / Avg: 235.77 / Max: 330 Min: 57.51 / Avg: 289.28 / Max: 672.46 Min: 32.28 / Avg: 274.76 / Max: 406.72 Min: 65.49 / Avg: 343.59 / Max: 811.63 Min: 21.48 / Avg: 206.2 / Max: 309.47 Min: 42.36 / Avg: 249.94 / Max: 651.86 Min: 71.79 / Avg: 281.33 / Max: 364.12 Min: 192.28 / Avg: 560.13 / Max: 714.79 Min: 72.24 / Avg: 295.54 / Max: 371.41 Min: 194.25 / Avg: 565.07 / Max: 686.05 Min: 55.23 / Avg: 273.91 / Max: 344.33 Min: 163.14 / Avg: 552.42 / Max: 653.46 Min: 33.09 / Avg: 260.62 / Max: 349.3 Min: 73.13 / Avg: 510.74 / Max: 686.96
OpenBenchmarking.org Watts, Fewer Is Better NAMD 3.0b6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17 / Avg: 109.51 / Max: 166.07 Min: 16.45 / Avg: 96.77 / Max: 141.98 Min: 27.01 / Avg: 187.68 / Max: 312.41 Min: 58.84 / Avg: 262.91 / Max: 648.62 Min: 29.28 / Avg: 180.85 / Max: 379.02 Min: 62.44 / Avg: 326.5 / Max: 779.77 Min: 20.68 / Avg: 165.67 / Max: 289.24 Min: 40.88 / Avg: 225.05 / Max: 615.88 Min: 73.54 / Avg: 248.99 / Max: 374.54 Min: 194.24 / Avg: 522.05 / Max: 693.75 Min: 73.99 / Avg: 269.88 / Max: 376.76 Min: 190.7 / Avg: 550.3 / Max: 686.19 Min: 55.01 / Avg: 258.53 / Max: 352.06 Min: 165.98 / Avg: 533.68 / Max: 633.47 Min: 33.34 / Avg: 236.13 / Max: 373.62 Min: 75.47 / Avg: 484.77 / Max: 671.95
OpenBenchmarking.org ns/day Per Watt, More Is Better NAMD 3.0b6 Input: ATPase with 327,506 Atoms EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0189 0.0378 0.0567 0.0756 0.0945 0.052 0.056 0.044 0.068 0.052 0.064 0.048 0.084 0.024 0.014 0.022 0.005 0.011 0.004 0.020 0.012
Result
OpenBenchmarking.org ns/day, More Is Better NAMD 3.0b6 Input: ATPase with 327,506 Atoms EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5 10 15 20 25 SE +/- 0.04338, N = 5 SE +/- 0.00431, N = 4 SE +/- 0.01368, N = 5 SE +/- 0.04751, N = 3 SE +/- 0.02148, N = 6 SE +/- 0.08771, N = 3 SE +/- 0.04917, N = 5 SE +/- 0.02125, N = 3 SE +/- 0.16058, N = 15 SE +/- 0.26203, N = 15 SE +/- 0.14537, N = 15 SE +/- 0.44989, N = 15 SE +/- 0.21567, N = 12 SE +/- 0.17435, N = 15 SE +/- 0.54361, N = 15 SE +/- 0.34017, N = 12 5.70672 5.44971 8.18275 17.81105 9.33534 20.75259 7.99756 18.97454 5.97130 7.30065 5.92501 2.56319 2.97021 2.21948 4.64814 5.85267 ns/day Per Watt
OpenBenchmarking.org ns/day Per Watt, More Is Better NAMD 3.0b6 Input: ATPase with 327,506 Atoms EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 0.0189 0.0378 0.0567 0.0756 0.0945 0.052 0.056 0.044 0.068 0.052 0.064 0.048 0.084 0.024 0.014 0.022 0.005 0.011 0.004 0.020 0.012 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better NAMD 3.0b6 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17 / Avg: 109.51 / Max: 166.07 Min: 16.45 / Avg: 96.77 / Max: 141.98 Min: 27.01 / Avg: 187.68 / Max: 312.41 Min: 58.84 / Avg: 262.91 / Max: 648.62 Min: 29.28 / Avg: 180.85 / Max: 379.02 Min: 62.44 / Avg: 326.5 / Max: 779.77 Min: 20.68 / Avg: 165.67 / Max: 289.24 Min: 40.88 / Avg: 225.05 / Max: 615.88 Min: 73.54 / Avg: 248.99 / Max: 374.54 Min: 194.24 / Avg: 522.05 / Max: 693.75 Min: 73.99 / Avg: 269.88 / Max: 376.76 Min: 190.7 / Avg: 550.3 / Max: 686.19 Min: 55.01 / Avg: 258.53 / Max: 352.06 Min: 165.98 / Avg: 533.68 / Max: 633.47 Min: 33.34 / Avg: 236.13 / Max: 373.62 Min: 75.47 / Avg: 484.77 / Max: 671.95
miniFE OpenBenchmarking.org Watts, Fewer Is Better miniFE 2.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 19.07 / Avg: 96.36 / Max: 147.72 Min: 15.16 / Avg: 86.83 / Max: 132.07 Min: 28.79 / Avg: 152.24 / Max: 300.13 Min: 53.09 / Avg: 309.09 / Max: 579.88 Min: 33.7 / Avg: 177.88 / Max: 364.34 Min: 63.04 / Avg: 330.55 / Max: 718.24 Min: 21.36 / Avg: 127.63 / Max: 261.53 Min: 40.05 / Avg: 258.88 / Max: 518.97 Min: 75.77 / Avg: 253.53 / Max: 365.64 Min: 189.32 / Avg: 498.77 / Max: 730.29 Min: 73.91 / Avg: 260.35 / Max: 366.7 Min: 199.32 / Avg: 500.61 / Max: 723.88 Min: 61.92 / Avg: 224.98 / Max: 377.92 Min: 168.8 / Avg: 424.3 / Max: 732.03 Min: 31.62 / Avg: 198.71 / Max: 331.13 Min: 73.08 / Avg: 366.83 / Max: 700.4
OpenBenchmarking.org CG Mflops Per Watt, More Is Better miniFE 2.2 Problem Size: Small EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90 180 270 360 450 254.52 282.72 334.41 140.84 298.90 179.90 392.87 162.67 79.24 39.85 71.52 43.11 156.59 110.48 160.18 114.81
Result
OpenBenchmarking.org CG Mflops, More Is Better miniFE 2.2 Problem Size: Small EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 13K 26K 39K 52K 65K SE +/- 49.20, N = 4 SE +/- 42.17, N = 4 SE +/- 42.89, N = 5 SE +/- 1510.37, N = 15 SE +/- 92.04, N = 5 SE +/- 2182.93, N = 15 SE +/- 53.07, N = 5 SE +/- 2087.13, N = 15 SE +/- 601.03, N = 12 SE +/- 1097.75, N = 15 SE +/- 572.00, N = 12 SE +/- 283.44, N = 3 SE +/- 11.14, N = 4 SE +/- 467.78, N = 4 SE +/- 135.77, N = 4 SE +/- 1633.92, N = 15 24526.0 24548.8 50909.6 43531.0 53167.8 59466.6 50139.9 42113.5 20091.1 19876.5 18621.2 21583.2 35230.1 46877.5 31830.2 42114.6 1. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi CG Mflops Per Watt
OpenBenchmarking.org CG Mflops Per Watt, More Is Better miniFE 2.2 Problem Size: Small EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 90 180 270 360 450 254.52 282.72 334.41 140.84 298.90 179.90 392.87 162.67 79.24 39.85 71.52 43.11 156.59 110.48 160.18 114.81 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better miniFE 2.2 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 130 260 390 520 650 Min: 19.07 / Avg: 96.36 / Max: 147.72 Min: 15.16 / Avg: 86.83 / Max: 132.07 Min: 28.79 / Avg: 152.24 / Max: 300.13 Min: 53.09 / Avg: 309.09 / Max: 579.88 Min: 33.7 / Avg: 177.88 / Max: 364.34 Min: 63.04 / Avg: 330.55 / Max: 718.24 Min: 21.36 / Avg: 127.63 / Max: 261.53 Min: 40.05 / Avg: 258.88 / Max: 518.97 Min: 75.77 / Avg: 253.53 / Max: 365.64 Min: 189.32 / Avg: 498.77 / Max: 730.29 Min: 73.91 / Avg: 260.35 / Max: 366.7 Min: 199.32 / Avg: 500.61 / Max: 723.88 Min: 61.92 / Avg: 224.98 / Max: 377.92 Min: 168.8 / Avg: 424.3 / Max: 732.03 Min: 31.62 / Avg: 198.71 / Max: 331.13 Min: 73.08 / Avg: 366.83 / Max: 700.4
RocksDB OpenBenchmarking.org Watts, Fewer Is Better RocksDB 9.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.34 / Avg: 155.39 / Max: 168.8 Min: 17.36 / Avg: 134 / Max: 149.2 Min: 24.6 / Avg: 300.55 / Max: 337.2 Min: 56.63 / Avg: 576.34 / Max: 690.47 Min: 34.3 / Avg: 383.63 / Max: 411.02 Min: 67.1 / Avg: 671.28 / Max: 820.11 Min: 22 / Avg: 301.4 / Max: 339.84 Min: 43.23 / Avg: 505.19 / Max: 682.09 Min: 72.31 / Avg: 332.58 / Max: 376.04 Min: 200.41 / Avg: 651.38 / Max: 755.47 Min: 74.89 / Avg: 331.82 / Max: 375.82 Min: 185.7 / Avg: 663.3 / Max: 752.58 Min: 52.65 / Avg: 330.54 / Max: 376.66 Min: 157.04 / Avg: 656.04 / Max: 755.91 Min: 33.11 / Avg: 323.9 / Max: 378.14 Min: 67.94 / Avg: 558.28 / Max: 752.93
OpenBenchmarking.org Op/s Per Watt, More Is Better RocksDB 9.0 Test: Read While Writing EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 14K 28K 42K 56K 70K 60067.21 66672.70 48605.69 34889.65 36906.83 29572.89 52342.46 45330.18 22369.26 14775.49 22200.62 16920.46 23723.76 18781.27 27560.28 22256.95
Result
OpenBenchmarking.org Op/s, More Is Better RocksDB 9.0 Test: Read While Writing EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 5M 10M 15M 20M 25M SE +/- 28098.62, N = 3 SE +/- 81998.56, N = 3 SE +/- 36524.26, N = 3 SE +/- 506458.47, N = 15 SE +/- 93669.97, N = 15 SE +/- 236709.61, N = 3 SE +/- 109424.14, N = 3 SE +/- 273109.47, N = 3 SE +/- 83772.27, N = 13 SE +/- 218479.19, N = 15 SE +/- 26369.56, N = 3 SE +/- 254193.37, N = 15 SE +/- 61518.27, N = 10 SE +/- 343737.14, N = 15 SE +/- 28201.44, N = 3 SE +/- 215352.04, N = 15 9333645 8934176 14608544 20108289 14158419 19851664 15776186 22900152 7439663 9624417 7366500 11223329 7841738 12321300 8926893 12425519 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better RocksDB 9.0 Test: Read While Writing EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 14K 28K 42K 56K 70K 60067.21 66672.70 48605.69 34889.65 36906.83 29572.89 52342.46 45330.18 22369.26 14775.49 22200.62 16920.46 23723.76 18781.27 27560.28 22256.95 CPU Power Consumption
OpenBenchmarking.org Watts, Fewer Is Better RocksDB 9.0 CPU Power Consumption Monitor EPYC 8534P EPYC 8534PN EPYC 9654 EPYC 9654 2P EPYC 9684X EPYC 9684X 2P EPYC 9754 EPYC 9754 2P Xeon Max 9468 Xeon Max 9468 2P Xeon Max 9480 Xeon Max 9480 2P Xeon Platinum 8490H Xeon Platinum 8490H 2P Xeon Platinum 8592+ Xeon Platinum 8592+ 2P 140 280 420 560 700 Min: 17.34 / Avg: 155.39 / Max: 168.8 Min: 17.36 / Avg: 134 / Max: 149.2 Min: 24.6 / Avg: 300.55 / Max: 337.2 Min: 56.63 / Avg: 576.34 / Max: 690.47 Min: 34.3 / Avg: 383.63 / Max: 411.02 Min: 67.1 / Avg: 671.28 / Max: 820.11 Min: 22 / Avg: 301.4 / Max: 339.84 Min: 43.23 / Avg: 505.19 / Max: 682.09 Min: 72.31 / Avg: 332.58 / Max: 376.04 Min: 200.41 / Avg: 651.38 / Max: 755.47 Min: 74.89 / Avg: 331.82 / Max: 375.82 Min: 185.7 / Avg: 663.3 / Max: 752.58 Min: 52.65 / Avg: 330.54 / Max: 376.66 Min: 157.04 / Avg: 656.04 / Max: 755.91 Min: 33.11 / Avg: 323.9 / Max: 378.14 Min: 67.94 / Avg: 558.28 / Max: 752.93
EPYC 8534P Processor: AMD EPYC 8534P 64-Core @ 2.30GHz (64 Cores / 128 Threads), Motherboard: AMD Cinnabar (RCB1009C BIOS), Chipset: AMD Device 14a4, Memory: 6 x 32GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92 + 2000GB Corsair MP700 PRO, Graphics: ASPEED, Network: 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00213Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 18 May 2024 13:07 by user phoronix.
EPYC 8534PN Processor: AMD EPYC 8534PN 64-Core @ 2.00GHz (64 Cores / 128 Threads), Motherboard: AMD Cinnabar (RCB1009C BIOS), Chipset: AMD Device 14a4, Memory: 6 x 32GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00213Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 19 May 2024 17:06 by user phoronix.
EPYC 9654 Processor: AMD EPYC 9654 96-Core @ 2.40GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 12 x 64GB DDR5-4800MT/s Samsung M321R8GA0BB0-CQKEG, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 13 May 2024 14:14 by user phoronix.
EPYC 9654 2P Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 14 May 2024 15:08 by user phoronix.
EPYC 9684X Processor: AMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 12 x 64GB DDR5-4800MT/s Samsung M321R8GA0BB0-CQKEG, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 May 2024 12:29 by user phoronix.
EPYC 9684X 2P Processor: 2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 11 May 2024 02:04 by user phoronix.
EPYC 9754 Processor: AMD EPYC 9754 128-Core @ 2.25GHz (128 Cores / 256 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 12 x 64GB DDR5-4800MT/s Samsung M321R8GA0BB0-CQKEG, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 15 May 2024 21:09 by user phoronix.
EPYC 9754 2P Processor: 2 x AMD EPYC 9754 128-Core @ 2.25GHz (256 Cores / 512 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 17 May 2024 00:34 by user phoronix.
Xeon Max 9468 Processor: Intel Xeon Max 9468 @ 3.50GHz (48 Cores / 96 Threads), Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS), Chipset: Intel Device 1bce, Memory: 576GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 21 April 2024 09:58 by user phoronix.
Xeon Max 9468 2P Processor: 2 x Intel Xeon Max 9468 @ 3.50GHz (96 Cores / 192 Threads), Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS), Chipset: Intel Device 1bce, Memory: 1136GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: 2 x Intel X710 for 10GBASE-T
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 22 April 2024 15:30 by user phoronix.
Xeon Max 9480 Processor: Intel Xeon Max 9480 @ 3.50GHz (56 Cores / 112 Threads), Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS), Chipset: Intel Device 1bce, Memory: 576GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 26 April 2024 09:38 by user phoronix.
Xeon Max 9480 2P Processor: 2 x Intel Xeon Max 9480 @ 3.50GHz (112 Cores / 224 Threads), Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS), Chipset: Intel Device 1bce, Memory: 1136GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: 2 x Intel X710 for 10GBASE-T
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 27 April 2024 16:18 by user phoronix.
Xeon Platinum 8490H Processor: Intel Xeon Platinum 8490H @ 3.50GHz (60 Cores / 120 Threads), Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS), Chipset: Intel Device 1bce, Memory: 512GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b000590Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 16 April 2024 00:42 by user phoronix.
Xeon Platinum 8490H 2P Processor: 2 x Intel Xeon Platinum 8490H @ 3.50GHz (120 Cores / 240 Threads), Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS), Chipset: Intel Device 1bce, Memory: 1008GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: 2 x Intel X710 for 10GBASE-T
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b000590Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 17 April 2024 09:38 by user phoronix.
Xeon Platinum 8592+ Processor: INTEL XEON PLATINUM 8592+ @ 3.90GHz (64 Cores / 128 Threads), Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS), Chipset: Intel Device 1bce, Memory: 512GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000200Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 11 April 2024 18:15 by user phoronix.
Xeon Platinum 8592+ 2P Processor: 2 x INTEL XEON PLATINUM 8592+ @ 3.90GHz (128 Cores / 256 Threads), Motherboard: Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS), Chipset: Intel Device 1bce, Memory: 1008GB, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: 2 x Intel X710 for 10GBASE-T
OS: Ubuntu 24.04, Kernel: 6.9.0-060900rc3-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000200Java Notes: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Notes: Python 3.12.2Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 8 April 2024 10:49 by user phoronix.