2024 Server CPUs

2 x Intel Xeon 6780E testing with a Quanta Cloud QuantaGrid D55Q-2U S7Q-MB-MPS-MDP (2A05.QCT001 BIOS) and ASPEED on Ubuntu 24.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2406049-NE-2024SERVE17
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Limit displaying results to tests within:

AV1 2 Tests
BLAS (Basic Linear Algebra Sub-Routine) Tests 4 Tests
C++ Boost Tests 5 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 8 Tests
C/C++ Compiler Tests 16 Tests
Compression Tests 3 Tests
CPU Massive 25 Tests
Creator Workloads 23 Tests
Cryptography 5 Tests
Database Test Suite 5 Tests
Encoding 7 Tests
Fortran Tests 7 Tests
Game Development 5 Tests
HPC - High Performance Computing 29 Tests
Imaging 3 Tests
Common Kernel Benchmarks 3 Tests
LAPACK (Linear Algebra Pack) Tests 3 Tests
Linear Algebra 2 Tests
Machine Learning 7 Tests
Molecular Dynamics 10 Tests
MPI Benchmarks 10 Tests
Multi-Core 44 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 7 Tests
OpenMPI Tests 18 Tests
Programmer / Developer System Benchmarks 12 Tests
Python 2 Tests
Raytracing 2 Tests
Renderers 5 Tests
Scientific Computing 15 Tests
Software Defined Radio 2 Tests
Server 7 Tests
Server CPU Tests 17 Tests
Single-Threaded 2 Tests
Video Encoding 7 Tests
Common Workstation Benchmarks 3 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable
Show Perf Per RAM Channel Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Xeon Platinum 8490H
April 16
  22 Hours, 19 Minutes
Xeon Platinum 8490H 2P
April 17
  23 Hours
Xeon Platinum 8592+
April 11
  22 Hours, 9 Minutes
Xeon Platinum 8592+ 2P
April 08
  23 Hours, 49 Minutes
Xeon Max 9468
April 21
  1 Day, 3 Hours, 43 Minutes
Xeon Max 9468 2P
April 22
  1 Day, 34 Minutes
Xeon Max 9480
April 26
  1 Day, 1 Hour, 37 Minutes
Xeon Max 9480 2P
April 27
  1 Day, 6 Hours, 26 Minutes
Xeon Platinum 8380 2P
May 07
  1 Day, 5 Hours, 29 Minutes
Xeon Platinum 8380
May 09
  1 Day, 7 Hours, 15 Minutes
EPYC 9684X 2P
May 11
  1 Day, 4 Hours, 57 Minutes
EPYC 9684X
May 12
  1 Day, 27 Minutes
EPYC 9654
May 13
  1 Day, 1 Hour, 48 Minutes
EPYC 9654 2P
May 14
  1 Day, 6 Hours, 48 Minutes
EPYC 9754
May 15
  1 Day, 6 Hours
EPYC 9754 2P
May 17
  1 Day, 9 Hours, 52 Minutes
EPYC 8534P
May 18
  1 Day, 3 Hours, 8 Minutes
EPYC 8534PN
May 19
  1 Day, 3 Hours, 48 Minutes
Xeon 6766E 2P
May 24
  1 Day, 3 Hours, 18 Minutes
Xeon 6766E
May 25
  18 Hours, 24 Minutes
Xeon 6780E
May 26
  17 Hours, 33 Minutes
Xeon 6780E 2P
May 27
  23 Hours, 13 Minutes
Invert Hiding All Results Option
  1 Day, 2 Hours, 10 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


2024 Server CPUsProcessorMotherboardChipsetMemoryDiskGraphicsNetworkMonitorOSKernelCompilerFile-SystemScreen ResolutionXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8380 2PXeon Platinum 8380EPYC 9684X 2PEPYC 9684XEPYC 9654EPYC 9654 2PEPYC 9754EPYC 9754 2PEPYC 8534PEPYC 8534PNXeon 6766E 2PXeon 6766EXeon 6780EXeon 6780E 2PIntel Xeon Platinum 8490H @ 3.50GHz (60 Cores / 120 Threads)Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS)Intel Device 1bce512GB2 x 1920GB KIOXIA KCD8XPUG1T92ASPEEDUbuntu 24.046.9.0-060900rc3-generic (x86_64)GCC 13.2.0ext41920x12002 x Intel Xeon Platinum 8490H @ 3.50GHz (120 Cores / 240 Threads)1008GB2 x Intel X710 for 10GBASE-TINTEL XEON PLATINUM 8592+ @ 3.90GHz (64 Cores / 128 Threads)512GB2 x INTEL XEON PLATINUM 8592+ @ 3.90GHz (128 Cores / 256 Threads)1008GB2 x Intel X710 for 10GBASE-TIntel Xeon Max 9468 @ 3.50GHz (48 Cores / 96 Threads)576GB2 x Intel Xeon Max 9468 @ 3.50GHz (96 Cores / 192 Threads)1136GB2 x Intel X710 for 10GBASE-TIntel Xeon Max 9480 @ 3.50GHz (56 Cores / 112 Threads)576GB2 x Intel Xeon Max 9480 @ 3.50GHz (112 Cores / 224 Threads)1136GB2 x Intel X710 for 10GBASE-T2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads)Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS)Intel Ice Lake IEH512GBVE2282 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP1920x1080Intel Xeon Platinum 8380 @ 3.40GHz (40 Cores / 80 Threads)256GB2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads)AMD Titanite_4G (RTI1007B BIOS)AMD Device 14a41520GBBroadcom NetXtreme BCM5720 PCIe1920x1200AMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads)12 x 64GB DDR5-4800MT/s Samsung M321R8GA0BB0-CQKEGAMD EPYC 9654 96-Core @ 2.40GHz (96 Cores / 192 Threads)2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads)1520GBAMD EPYC 9754 128-Core @ 2.25GHz (128 Cores / 256 Threads)12 x 64GB DDR5-4800MT/s Samsung M321R8GA0BB0-CQKEG2 x AMD EPYC 9754 128-Core @ 2.25GHz (256 Cores / 512 Threads)1520GBAMD EPYC 8534P 64-Core @ 2.30GHz (64 Cores / 128 Threads)AMD Cinnabar (RCB1009C BIOS)6 x 32GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG2 x 1920GB KIOXIA KCD8XPUG1T92 + 2000GB Corsair MP700 PRO2 x Broadcom NetXtreme BCM5720 PCIeAMD EPYC 8534PN 64-Core @ 2.00GHz (64 Cores / 128 Threads)2 x 1920GB KIOXIA KCD8XPUG1T922 x Intel Xeon 6766E @ 2.70GHz (288 Cores)Quanta Cloud QuantaGrid D55Q-2U S7Q-MB-MPS-MDP (2A05.QCT001 BIOS)Intel Ice Lake IEH512GB3201GB Micron_7450_MTFDKCB3T2TFS + 2 x 1920GB KIOXIA KCD8XPUG1T922 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMAIntel Xeon 6766E @ 2.70GHz (144 Cores)256GB2 x 1920GB KIOXIA KCD8XPUG1T92 + 3201GB Micron_7450_MTFDKCB3T2TFSIntel Xeon 6780E @ 3.00GHz (144 Cores)3201GB Micron_7450_MTFDKCB3T2TFS + 2 x 1920GB KIOXIA KCD8XPUG1T922 x Intel Xeon 6780E @ 3.00GHz (288 Cores)512GBOpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --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 Details- Xeon Platinum 8490H: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b000590- Xeon Platinum 8490H 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b000590- Xeon Platinum 8592+: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000200- Xeon Platinum 8592+ 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000200- Xeon Max 9468: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290- Xeon Max 9468 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290- Xeon Max 9480: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290- Xeon Max 9480 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290- Xeon Platinum 8380 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0xd0003d1- Xeon Platinum 8380: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0xd0003d1- EPYC 9684X 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144- EPYC 9684X: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144- EPYC 9654: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144- EPYC 9654 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101144- EPYC 9754: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116- EPYC 9754 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116- EPYC 8534P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00213- EPYC 8534PN: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00213- Xeon 6766E 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x13000131- Xeon 6766E: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x13000131- Xeon 6780E: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x13000131- Xeon 6780E 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x13000131Java Details- OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Details- Python 3.12.2Security Details- Xeon Platinum 8490H: 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 Platinum 8490H 2P: 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 Platinum 8592+: 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 Platinum 8592+ 2P: 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: 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: 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 9480: 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 9480 2P: 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 Platinum 8380 2P: gather_data_sampling: Mitigation of Microcode + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + 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 Platinum 8380: gather_data_sampling: Mitigation of Microcode + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + 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- EPYC 9684X 2P: 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 9684X: 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: 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: 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 9754: 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 9754 2P: 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 8534P: 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: 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- Xeon 6766E 2P: 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: Not affected + srbds: Not affected + tsx_async_abort: Not affected- Xeon 6766E: 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: Not affected + srbds: Not affected + tsx_async_abort: Not affected- Xeon 6780E: 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: Not affected + srbds: Not affected + tsx_async_abort: Not affected- Xeon 6780E 2P: 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: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Xeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8380 2PXeon Platinum 8380EPYC 9684X 2PEPYC 9684XEPYC 9654EPYC 9654 2PEPYC 9754EPYC 9754 2PEPYC 8534PEPYC 8534PNXeon 6766E 2PXeon 6766EXeon 6780EXeon 6780E 2PLogarithmic Result OverviewPhoronix Test SuiteNAMDStockfishASTC EncoderRocksDBCoremarkJohn The RipperminiBUDEOpenSSLSpeedbBlenderACES DGEMMHelsingPrimesievem-queensOSPRayOSPRay StudioLlamafileMemcached7-Zip CompressionOpenVKLOpenVINOEmbreePyTorchLULESHoneDNNONNX RuntimeTimed Node.js CompilationKvazaarTimed LLVM CompilationLuxCoreRenderPostgreSQLminiFEParallel BZIP2 CompressionIntel Open Image Denoiseuvg266QuantLibTimed Linux Kernel CompilationGraphicsMagickCloverLeafY-CruncherLiquid-DSPsrsRAN ProjectNumpy BenchmarkTensorFlowFFmpegx265ClickHouseBRL-CADXmriglibavif avifencTimed FFmpeg CompilationTimed Godot Game Engine CompilationGNU Octave BenchmarkApache IoTDBSVT-AV1Timed ImageMagick CompilationTimed Gem5 CompilationZstd CompressionWebP Image EncodePyBenchGoogle SynthMarkTimed Mesa CompilationSecureMark

Xeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8380 2PXeon Platinum 8380EPYC 9684X 2PEPYC 9684XEPYC 9654EPYC 9654 2PEPYC 9754EPYC 9754 2PEPYC 8534PEPYC 8534PNXeon 6766E 2PXeon 6766EXeon 6780EXeon 6780E 2PLogarithmic Per Watt Result OverviewPhoronix Test SuiteminiFEMemcachedWebP Image EncodeZstd CompressionSVT-AV1PyTorchFFmpegASTC EncodersrsRAN ProjectNumpy Benchmarkx265Google SynthMarkminiBUDESecureMarkCoremarkACES DGEMM7-Zip CompressionKvazaarClickHouseXmrigSpeedbTensorFlowuvg266LuxCoreRenderRocksDBGraphicsMagickLiquid-DSPEmbreeOpenVKLQuantLibOSPRayLULESHJohn The RipperStockfishBRL-CADOpenSSLIntel Open Image DenoiseLlamafileNAMDP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.M

2024 Server CPUswrf: conus 2.5kmopenvkl: vklBenchmarkCPU ISPCrelion: Basic - CPUnwchem: C240 Buckyballhpcg: 144 144 144 - 60tensorflow: CPU - 512 - ResNet-50brl-cad: VGR Performance Metricincompact3d: X3D-benchmarking input.i3dcloverleaf: clover_bm16askap: tConvolve MT - Griddingaskap: tConvolve MT - Degriddingpytorch: CPU - 256 - ResNet-152pgbench: 100 - 1000 - Read Only - Average Latencypgbench: 100 - 1000 - Read Onlytensorflow: CPU - 256 - ResNet-50openssl: RSA4096openssl: RSA4096build-linux-kernel: allmodconfigstockfish: Chess Benchmarkapache-iotdb: 800 - 100 - 800 - 100apache-iotdb: 800 - 100 - 800 - 100pytorch: CPU - 512 - ResNet-152pytorch: CPU - 64 - ResNet-152securemark: SecureMark-TLSclickhouse: 100M Rows Hits Dataset, Third Runclickhouse: 100M Rows Hits Dataset, Second Runclickhouse: 100M Rows Hits Dataset, First Run / Cold Cacheapache-iotdb: 500 - 100 - 800 - 400apache-iotdb: 500 - 100 - 800 - 400openradioss: Chrysler Neon 1Mnamd: STMV with 1,066,628 Atomsbuild-gem5: Time To Compilelammps: 20k Atomsbuild-nodejs: Time To Compileopenssl: SHA256tensorflow: CPU - 512 - GoogLeNetffmpeg: libx265 - Platformffmpeg: libx265 - Video On Demandonnx: fcn-resnet101-11 - CPU - Standardonnx: fcn-resnet101-11 - CPU - Standardffmpeg: libx265 - Uploadpgbench: 100 - 1000 - Read Write - Average Latencypgbench: 100 - 1000 - Read Writeblender: Barbershop - CPU-Onlynumpy: luxcorerender: DLSC - CPUopenssl: AES-128-GCMopenssl: AES-256-GCMopenssl: ChaCha20openssl: SHA512openssl: ChaCha20-Poly1305pytorch: CPU - 512 - ResNet-50luxcorerender: LuxCore Benchmark - CPUpytorch: CPU - 1 - ResNet-152xmrig: GhostRider - 1Mluxcorerender: Orange Juice - CPUospray: particle_volume/scivis/real_timerocksdb: Read While Writingonnx: Faster R-CNN R-50-FPN-int8 - CPU - Standardonnx: Faster R-CNN R-50-FPN-int8 - CPU - Standardopenfoam: drivaerFastback, Medium Mesh Size - Execution Timeopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timebuild-llvm: Ninjamemcached: 1:100onnx: ResNet50 v1-12-int8 - CPU - Standardonnx: ResNet50 v1-12-int8 - CPU - Standardnamd: ATPase with 327,506 Atomsspeedb: Read While Writingpytorch: CPU - 64 - ResNet-50onnx: ArcFace ResNet-100 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardonnx: bertsquad-12 - CPU - Standardonnx: bertsquad-12 - CPU - Standardonnx: CaffeNet 12-int8 - CPU - Standardonnx: CaffeNet 12-int8 - CPU - Standardllamafile: wizardcoder-python-34b-v1.0.Q6_K - CPUbuild-godot: Time To Compilepytorch: CPU - 256 - ResNet-50openradioss: INIVOL and Fluid Structure Interaction Drop Containerqmcpack: Li2_STO_aeonnx: fcn-resnet101-11 - CPU - Parallelonnx: fcn-resnet101-11 - CPU - Parallelonnx: CaffeNet 12-int8 - CPU - Parallelonnx: CaffeNet 12-int8 - CPU - Parallelonnx: T5 Encoder - CPU - Standardonnx: T5 Encoder - CPU - Standardopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP16 - CPUonnx: GPT-2 - CPU - Standardonnx: GPT-2 - CPU - Standardtensorflow: CPU - 256 - GoogLeNetospray: particle_volume/ao/real_timeonnx: yolov4 - CPU - Standardonnx: yolov4 - CPU - Standardonnx: super-resolution-10 - CPU - Standardonnx: super-resolution-10 - CPU - Standardospray-studio: 1 - 4K - 32 - Path Tracer - CPUospray-studio: 3 - 4K - 32 - Path Tracer - CPUcompress-zstd: 19 - Decompression Speedcompress-zstd: 19 - Compression Speedspeedb: Rand Readvvenc: Bosphorus 4K - Fastcompress-zstd: 19, Long Mode - Decompression Speedcompress-zstd: 19, Long Mode - Compression Speedopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUospray-studio: 1 - 4K - 1 - Path Tracer - CPUtensorflow: CPU - 1 - ResNet-50tensorflow: CPU - 64 - ResNet-50openvino: Machine Translation EN To DE FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUonnx: ResNet50 v1-12-int8 - CPU - Parallelonnx: ResNet50 v1-12-int8 - CPU - Parallelopenvino: Noise Suppression Poconet-Like FP16 - CPUopenvino: Noise Suppression Poconet-Like FP16 - CPUavifenc: 0onnx: T5 Encoder - CPU - Parallelonnx: T5 Encoder - CPU - Parallelaskap: tConvolve MPI - Griddingaskap: tConvolve MPI - Degriddinghelsing: 14 digitospray-studio: 1 - 4K - 16 - Path Tracer - CPUospray-studio: 3 - 4K - 16 - Path Tracer - CPUcloverleaf: clover_bm64_shortospray-studio: 3 - 4K - 1 - Path Tracer - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUonnx: ArcFace ResNet-100 - CPU - Parallelonnx: ArcFace ResNet-100 - CPU - Parallelopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUonnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallelonnx: Faster R-CNN R-50-FPN-int8 - CPU - Paralleltensorflow: CPU - 512 - AlexNetquantlib: Multi-Threadedopenvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUospray: gravity_spheres_volume/dim_512/pathtracer/real_timeblender: Pabellon Barcelona - CPU-Onlyopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUonnx: bertsquad-12 - CPU - Parallelonnx: bertsquad-12 - CPU - Parallelopenvino: Person Re-Identification Retail FP16 - CPUopenvino: Person Re-Identification Retail FP16 - CPUpytorch: CPU - 1 - ResNet-50openvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUonnx: GPT-2 - CPU - Parallelonnx: GPT-2 - CPU - Parallelopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUonnx: yolov4 - CPU - Parallelonnx: yolov4 - CPU - Parallelgraphics-magick: Noise-Gaussiangpaw: Carbon Nanotubegraphics-magick: Sharpenbuild-linux-kernel: defconfiggraphics-magick: Enhancedonnx: super-resolution-10 - CPU - Parallelonnx: super-resolution-10 - CPU - Parallelgraphics-magick: Swirlrocksdb: Rand Readminibude: OpenMP - BM2minibude: OpenMP - BM2coremark: CoreMark Size 666 - Iterations Per Secondsrsran: PDSCH Processor Benchmark, Throughput Totalblender: Classroom - CPU-Onlyaircrack-ng: webp: Quality 100, Lossless, Highest Compressionavifenc: 2srsran: PUSCH Processor Benchmark, Throughput Totalllamafile: llava-v1.5-7b-q4 - CPUlulesh: compress-7zip: Decompression Ratingcompress-7zip: Compression Ratingamg: liquid-dsp: 1 - 256 - 512npb: EP.Dastcenc: Very Thoroughtensorflow: CPU - 256 - AlexNetminife: Smallopenradioss: Cell Phone Drop Testvvenc: Bosphorus 4K - Fasterprimesieve: 1e13gromacs: MPI CPU - water_GMX50_bareastcenc: Exhaustiveuvg266: Bosphorus 4K - Mediumjohn-the-ripper: WPA PSKtensorflow: CPU - 1 - GoogLeNetliquid-dsp: 1 - 256 - 57quantlib: Single-Threadedliquid-dsp: 64 - 256 - 512svt-av1: Preset 13 - Bosphorus 4Kuvg266: Bosphorus 4K - Slowliquid-dsp: 1 - 256 - 32tensorflow: CPU - 64 - GoogLeNetliquid-dsp: 64 - 256 - 57liquid-dsp: 128 - 256 - 512liquid-dsp: 256 - 256 - 512john-the-ripper: bcryptliquid-dsp: 128 - 256 - 57svt-av1: Preset 4 - Bosphorus 4Kembree: Pathtracer ISPC - Asian Dragon Objjohn-the-ripper: Blowfishliquid-dsp: 256 - 256 - 57liquid-dsp: 256 - 256 - 32liquid-dsp: 128 - 256 - 32synthmark: VoiceMark_100liquid-dsp: 64 - 256 - 32uvg266: Bosphorus 4K - Very Fastblender: Junkshop - CPU-Onlyblender: Fishy Cat - CPU-Onlyappleseed: Disney Materialtensorflow: CPU - 64 - AlexNetkvazaar: Bosphorus 4K - Mediumpennant: sedovbigkvazaar: Bosphorus 4K - Slowsrsran: PUSCH Processor Benchmark, Throughput Threadkvazaar: Bosphorus 4K - Very Fastopenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenfoam: drivaerFastback, Small Mesh Size - Mesh Timesvt-av1: Preset 8 - Bosphorus 4Koidn: RTLightmap.hdr.4096x4096 - CPU-Onlyblender: BMW27 - CPU-Onlyminibude: OpenMP - BM1minibude: OpenMP - BM1x265: Bosphorus 4Kbuild-ffmpeg: Time To Compilepennant: leblancbigtensorflow: CPU - 1 - AlexNetnpb: SP.Cpybench: Total For Average Test Timesnpb: LU.Castcenc: Thoroughnpb: IS.Dbuild-mesa: Time To Compileincompact3d: input.i3d 193 Cells Per Directiony-cruncher: 1Bbuild-imagemagick: Time To Compilem-queens: Time To Solveoidn: RT.ldr_alb_nrm.3840x2160 - CPU-Onlyoidn: RT.hdr_alb_nrm.3840x2160 - CPU-Onlynpb: CG.Coctave-benchmark: y-cruncher: 500Membree: Pathtracer ISPC - Asian Dragonwebp: Quality 100, Highest Compressionembree: Pathtracer ISPC - Crownastcenc: Mediumsvt-av1: Preset 12 - Bosphorus 4Kavifenc: 6, Losslessmt-dgemm: Sustained Floating-Point Rateavifenc: 10, Losslessnpb: MG.Csrsran: PDSCH Processor Benchmark, Throughput Threadlammps: Rhodopsin Proteinavifenc: 6primesieve: 1e12compress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressiononednn: Deconvolution Batch shapes_3d - CPUXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8380 2PXeon Platinum 8380EPYC 9684X 2PEPYC 9684XEPYC 9654EPYC 9654 2PEPYC 9754EPYC 9754 2PEPYC 8534PEPYC 8534PNXeon 6766E 2PXeon 6766EXeon 6780EXeon 6780E 2P11980.8061734373.1822070.331.0738148.45833892357.009712328.478088.1120.250.5721748146143.67974189.927431.1362.4549688747985.169087177020.4620.13277205481.66486.05465.50294.3387797206207.201.05274210.93434.892254.00361806994523461.2349.7949.84116.3158.5971924.9714.04071227261.97455.578.727688046936977002393527173161264211732221992772322503885073056.447.1019.415808.813.4621.8932784173829.408034.0021420.6467160.72685187.0174481667.953.17002315.4102.970211069536056.0118.931952.816749.637620.14531.17292852.3265.85140.23354.84140.43109.30329.0363.039221.27368784.3042.79694357.355100.39298.564.49976222.029448.4621.932450.352219.85934.39649227.42358026680061032.014.62761796176.7191069.17.940.43115106.30165710.53120.9362.46477.773.19298313.10013.954209.1069.9313.06500326.14321314.217891.693.348313803664939.48197755.991068.4321.343646.849136.201643.4317.843325.9030.916532.34331278.45179647.5356.38167.8324.644587.9216.503606.1262.127616.09569.795943.6651.463.5815943.175.21520191.5626.439004.2355.120418.141315847.77615739.4502164.86604205.52550226536617086.9672174.1922194630.60044924688.568.61147469.3280.5237.7522616.920.9523072.3022878623881901605097000168823337562.807.08851205.3035230.131.4915.06051.8208.7624.268917.8540127833.56670746673209.6744410000162.47415.7136506000380.14253920000091856666710208333339455131812666677.06179.380594513350083333332438000003200600000711.378189056666749.4235.9634.8353.490276969.6220.9716.0308020.23129.750.0342.21941931.38458570.9081.2925.8982.8482071.18937.8624.0564.95586556.5371245.26765135125.1551.29712865.7219.22612.92934137.04714.72415.2112.712.7133811.637.5323.26791.99253.0870.0540333.9902164.2506.27823.5437054.92383448.70680.228.7163.1694.5762.1448571.077436674.122716203.4071960.960.8515174.954136969196.808116214.807630.0614.081.293778141161.061916162.954601.1218.57018776176992.238365178514.5914.03275472501.63507.56485.63317.3680169620115.891.10900178.28250.521172.990119106802107576.0848.7648.7792.347310.8362324.5614.85667315141.54456.1811.12153302630072713695343284706244029729474284625447744220253297737.558.8215.7711263.218.7038.22691232130029.753633.6082196.49644162.02669122.3143245908.405.62517177.7632.219481729702737.0827.594436.249360.358916.57051.43573696.9803.82118.57237.44107.36101.629352.7382.834981.76353566.3773.46571288.47145.78654.845.55101180.032528.9338.223564.285815.55554.72530212.9243341138813963.615.25512869586.480992.18.160.29189893.318944.39115.4737.58793.646.39031156.46115.967356.6367.3754.05881246.33541773.034614.150.831141381683726.58106365.021843.6134.305429.148725.124744.0521.735483.8133.289930.03681899.64350187.5283.11423.1842.818846.3815.047862.5868.829914.52807.7215519.9235.532.5047654.956.90619144.6405.7220937.5878.398912.754618241.56326529.9813425.29287188.890617525049998140.8283520.7053902076.24361230108.735.9153914.9330.5137.3114853.610.4443724.06255705459272931244026671684333314946.7613.75181730.2546877.528.2212.53627.10715.4508.370127.6579325313.50669913333210.1799496667154.50224.7136461000331.7626555000001126466667161696666718360337236666676.698132.2806183583536126666761165666673558600000712.697202500000054.8821.1519.1348.56418996.4732.0411.2843831.24129.767.6428.55603735.60409765.2242.1813.74133.6283340.69437.0419.5062.52980737.57147311.14767261127.0986.98453960.2818.2936.391698175.00813.2627.8814.484.4158128.248.0362.364152.74723.07124.7084484.5499156.5166.17245.9659555.074163333.21675.542.2813.0442.3251.4692100.73617010602.4082004290.4481878.335.1361140.53916904323.245697410.595062.0723.330.5891699046129.021036735.631594.4316.90210976634581.739461680723.4123.49309896458.23455.79437.20280.2692709751157.561.43787187.50737.460214.43166573957920462.7853.9253.99109.6709.1188526.9313.26475396236.73520.229.388895675065407443876836703441509169372388605477022362573801363.308.0627.218378.314.7430.6799892689329.018534.4595302.0641148.40248165.0484752673.793.18392314.0574.648141216841363.0615.037666.516239.745125.15911.10042908.1874.08122.71863.60128.74105.98302.2663.308361.20298830.7252.16966460.77539.62403.203.69649270.329423.5930.840443.535522.96893.85774259.18650798595221221.818.03151783817.3921242.69.800.4981279.7914416.4391.5126.85592.203.23780308.8019.406757.2664.4752.15980462.85537265.931177.289.070231253199339.38170955.071159.6616.553460.409421.872924.6613.734656.9229.956633.37971486.12195648.8224.64284.6427.101080.2911.175725.0350.830719.67736.729515.8677.522.3626943.424.01687248.6825.1912317.9447.895320.879517143.04816734.5912284.26352234.540529297517946108.1242703.0882369998.35694026378.163.61153584.2030.6034.7913088.716.6138433.9733381464416361814760000187743338089.147.62401277.4631830.227.9015.96648.40610.4774.630316.6442811621.67731406673600.5783770000174.83216.0640244000319.7624755000001152933333117876666710578135032000007.87697.1861105831380683333331672000003148233333799.717192776666735.1032.0831.9148.425051819.8924.8016.9691024.23129.538.0233.06539331.70490579.4371.5823.72106.3952659.88338.9321.3574.18270735.50109416.42691232012.7754.45474059.7017.25810.19670666.70112.59514.2573.323.3163760.456.4583.391112.66113.4782.9663327.7829177.8615.78429.1410784.456107321.89742.934.3012.9714.2501.8475370.9071615601.1013222155.7411809.769.2976169.755186677171.375778378.056038.9216.001.149881881150.332057021.661499.6189.73922261037882.739277163916.1716.25308434438.60443.58423.30286.788923185086.932.25422157.36356.400145.413131976050120577.7653.2253.29100.7389.9268426.8030.20333110125.87519.8912.74176931314601314975299393776842843827604652076070744409914427040.838.8817.5716324.719.9153.93481242551928.848134.6628137.03802139.61713106.8153780240.565.83954171.2315.852671758573740.7726.086638.424044.599622.42511.32454754.8053.05103.37641.45100.0097.911327.8753.050851.64919605.5892.80263356.76443.41736.234.29531236.648501.3954.119558.939416.96723.98013251.30524249333911165.817.85861478716.9601189.79.670.46128566.817593.8594.6628.171128.786.49693153.92213.159360.8960.5262.81739354.75472004.059139.248.091121241435831.9089857.132237.6931.790731.458922.965570.6314.678719.3731.532431.71092039.63380867.5234.24545.2148.757641.8511.6910934.1764.789115.43416.9518375.6345.692.4848122.165.10948195.4535.1824676.8670.437714.198019438.04928425.8723744.65431214.812689578392800191.8184795.4624966250.77798330547.633.6164181.2930.5933.6505493.47.4970826.73457341067423936096463331820053316004.0714.63341681.4442114.626.8813.25625.02118.5408.909130.3779709312.69704990003586.3883701250162.97927.1138950600291.3726200666671493833333202700000020677844505250007.360173.2208205499628430000060773000003520766667799.491215463333353.6618.3117.2943.695399858.8441.589.96452640.92129.565.3823.55893229.32512570.4262.4012.76199.7344993.34339.3117.3892.15945930.76228184.68692455312.0997.93837437.8116.1835.391259675.11011.4937.4025.095.0398764.936.8562.659199.60893.46153.1622550.0168165.4955.61358.2533404.579220893.78755.555.4962.8052.1871.4323620.62425013273.6941243412.3712530.525.2423117.53713937407.589406351.293375.3219.200.5591792636119.73697707.222861.0428.3876801908890.028574454219.4719.01277647470.02465.06446.40314.4482455664228.741.45223220.56928.278270.38044439185187383.5747.3247.34139.0967.2418723.3314.29269969320.11461.617.466412008481705074567280872265973541531619926127716132105821351.056.3019.995446.811.8616.0869743966333.334129.9973440.83113153.12467212.4824384040.523.02897330.1045.97130879540851.5521.217947.270347.241721.17021.37831728.4691.47149.37852.97160.56121.19380.9022.625441.45280687.6714.57581219.53241.93285.955.67700176.333375.7316.050254.959918.21794.58912217.88068973807421126.414.42114570486.3241131.17.760.33135075.07198811.29104.2328.73415.963.13366319.04311.374208.4976.9152.97094337.00421766.518522123.832370554283741.43235755.74860.6424.553540.728923.122074.3016.332935.4735.066528.51641047.96143755.3236.52202.5919.9102106.9612.543802.6161.254716.32517.136717.6550.992.5019183.835.22781191.1115.279087.1261.825616.174215054.37611443.4211605.10914195.70539420534615470.2131755.3331614763.70323721286.686.67117588.9920.5541.1252170.77.6723848.5402077383147161588496667169063335366.695.1342986.0420091.133.7514.10167.3247.5453.120415.1728655236.13671303333225.7632396667144.69813.3636537667333.7321387666678001866678385300006787526700333336.49064.959767893284893333323818000002351900000711.101164026666739.5642.8542.7858.379116841.6818.1321.0211217.52129.740.2345.94743531.78187462.7501.0831.6663.6201590.50129.0526.4585.93095748.2858815.36763126171.5237.00513050.1620.17513.96923768.30016.07521.0302.272.2635116.557.6223.78175.46903.1058.4230247.2979146.0976.83619.7689304.91685910.95702.627.6293.7306.1602.6320771.264077157.9611776214.416178554.4054145.293658790189.686116263.333347.9316.301.027976035141.951395446.445531.0246.93613797720394.848077046516.6815.84277589501.07499.18491.89328.4177501248125.151.88679184.29846.705175.62588635753413467.9145.9545.29121.6708.3996222.3416.96358952169.44435.2110.17127758338275710118989933234499084949273230422299332039672044741.656.8117.9110975.816.9129.3496962441736.228327.6074201.10956134.78213134.4193373323.925.74491174.6777.300651385925442.3524.651040.567160.112016.63941.65270604.7551.77123.88241.68114.29102.12460.6732.174632.02659492.8855.02267199.07942.16568.817.04111141.954447.5029.501869.450114.48314.85818205.88538926452111040.213.84199435706.0231041.87.620.28174862.5610495.11111.7129.98795.636.00525166.48517.155565.2369.6474.05270246.68042939.236900.565.026166451970132.25123756.711691.7930.473332.814124.003998.0217.195576.8640.453424.72741604.11288129.1256.63373.5035.932455.2613.347127.2872.801813.73667.3213102.2846.492.5337710.607.34848135.9765.5517252.2776.422713.086915042.87319932.1432695.74921173.909509405533173127.8453196.1232993136.60183230236.044.7183665.7580.5438.6533787.07.7752102.66840170749172931562616671690466710970.4510.20511479.5319876.527.5111.89733.92113.3926.212622.2556287518.24670640003207.0732560000122.61719.7536497667341.5124940500001091460000144406666713386635964000005.991104.8425134562482513333346709666673254033333710.916196323333344.7524.9923.1751.245898932.5926.6913.2112526.00129.751.9128.99311932.48199354.8231.7216.55108.3972709.93027.2620.7882.88109937.76135939.25763246822.5272.32973816.7018.4747.293491176.33414.40210.8403.493.5165396.2212.6173.110125.74943.09107.6299441.6518122.0897.12139.3457325.317172614.10651.947.5123.7763.1211.9829320.94507512410.291356384.3842245.230.5872125.14767812351.644165357.573406.9619.580.6221607424123.69813667.625441.5384.0637359223089.538635451819.2919.24277260476.91474.68461.62309.0383475645212.991.50214215.49632.157252.64851840436763398.4048.3548.14155.7886.5496523.7713.84572225290.56464.728.007477107026605920141154532637406900671889398078318779617444051.866.8218.076122.312.5018.4851736650033.987729.4250410.29005150.66848197.8444315467.173.37034296.7995.92501998817651.3523.662542.274749.974020.01031.41662705.6061.54143.80851.56146.03117.81382.1092.617211.47249678.4834.64757215.12759.69301.305.91961168.859386.0018.553359.179516.91664.67269213.98562540733661127.514.52440706746.5011133.37.720.37140144.0117929.99103.5638.50467.083.32030301.13612.724386.4873.0602.88734346.21323250.220561.4107.237337903923842.22212362.53894.8723.656542.271124.942243.8419.442877.3235.111028.47981108.95162727.4257.50217.1122.362597.8013.594093.7363.535315.73947.767207.7448.592.6720910.535.29027188.9305.779686.3961.010716.389815848.48713040.5011815.26075190.06343223446993978.1051952.6061855135.32609815081.076.87134048.9640.5539.8762514.08.1325831.8382433473423631597993333168853336445.095.97021058.9918621.231.9214.36258.1598.2823.627615.6833407631.47671276673229.1689003333149.99613.8936539000335.5823089666678822000009445500007914429724000006.57271.825379152322253333327690666672739633333712.013167970000040.8239.1738.4856.401698852.3618.8419.8824618.24129.741.3441.74619231.99824865.3591.1428.4970.8431771.06531.4024.8405.11416143.6769776.41762131606.8642.98343076.6619.65012.79975827.95915.37618.1022.392.3835234.067.7483.70483.49213.1065.3318284.3336151.7186.60522.5399904.86786407.46705.530.8433.5505.3322.2877661.196436840.0571852201.3821873.162.6357138.394075342187.224004269.204382.9714.921.098920725135.281627941.448534.8231.03314940908295.398013956614.7414.96277298471.79467.96460.66325.9778057728117.911.76778181.62951.816171.174103382860113477.3346.5946.79121.7808.2601623.3016.65560046156.70439.8410.44148466962705011811489573535229124033473756452205337273521529737.847.4114.8612115.016.6433.12751122332936.003927.7772186.32539159.31171128.0663392842.386.17796161.8862.563191535440038.3029.444833.962155.901917.98731.71875581.7691.54122.70437.91110.56106.27475.2542.104802.08780478.4675.39123185.4664.91569.607.53259132.712449.9133.112569.201414.45835.00222199.89636179419711039.213.34979074906.0771046.07.460.30176900.329614.37107.3041.57889.186.51744153.40718.725933.6568.4373.95819252.67145195.140368.658.248153571811132.44114362.111801.8934.250729.195525.724351.5920.245525.5640.517824.68181677.62316359.2278.52401.4539.338951.2914.427689.3273.968413.51907.9714033.9435.462.7940032.817.30822136.6896.0218561.9179.377012.597416443.91122730.5303015.88628169.850558473755946133.9903349.7563279262.53034911866.740.3865663.8230.5438.1364386.67.2653269.46946913152551432051610001688400012793.0111.88751528.4121583.227.6311.96730.19614.1277.207524.6565788914.67662660003222.8757176667124.26322.1436521333308.9524913400001206066667151893333315595938965000006.052112.7231156658503600000053670333333337066667712.085209046666746.2523.4521.4549.136175940.4029.6212.1069028.96129.752.2327.57729128.84015755.9791.7415.29114.9022872.56227.9020.0082.52736333.21141954.15763257384.5583.33953819.8318.1386.663875106.26614.0949.2853.663.6566492.4412.5782.961129.74823.09116.6053483.1049127.5467.03844.6683025.311171436.47663.852.2443.7382.6911.7112960.86435814030.5131527368.219217030.1110127.972300947292.030314295.305279.079893.3214.880.5251930006125.911099481.536371.6356.987104597200102.677517848014.7615.11231076295.06295.43285.17349.4274653007185.031.61232219.66935.701303.42154530700576411.0543.5543.80110.9709.0336721.7313.69373042235.31394.718.988072292653007124854114474254511900632207953990029350212241337.456.4316.685230.914.3117.6222983668929.764533.6027362.5935162.32335194.3824814941.534.49317222.5226.726631179531837.7521.956445.541454.283218.42221.42849701.7243.10151.53736.97135.56131.37345.9382.892531.55821641.0013.43125292.517113.70175.725.16480194.149402.7817.826960.928216.41235.29844192.8105524764188900.514.32591364665.877912.97.830.45151627.4215539.13108.3270.95281.365.18070192.99326.332963.8288.0654.00675249.48414497.312800.482.959304423486635.29183316.841186.2425.981338.502560.901313.394.544390.8138.517925.96061470.46182366.9207.8496.0822.508176.109.822030.9766.913614.946024.393263.0342.648.279657.086.50747153.5385.9313467.2870.340914.216915348.45314439.4102155.59763178.61540325123410397.0752426.8811839444.39631118943.662.54159709.7470.4947.5053288.412.2921389.823366534391719723153533135990008880.824.29351340.3431046.531.1713.02340.4448.9642.647116.1949158329.18539303332558.6646513333138.72213.7931992000347.721999266667981090000106103333311321926890333335.43784.8010113223303710000035722333333044700000548.362173116666740.6933.2030.7558.376732904.3620.3315.2471919.71112.344.1140.28640436.86211659.0971.4823.6785.2892132.23530.8025.9394.70255737.2690420.25971185954.5430.03093345.4521.43711.24980167.74416.33412.3903.103.0838906.139.6943.72499.24302.8182.9195216.9678140.7567.38824.9741045.919110742.15490.333.5443.5283.3662.1814550.86706518998.816869705.5843576.420.412991.75430885558.384868512.365172.967425.2217.700.581172076291.65587354.818389.3612.78251099545100.297672949517.5517.73231357342.99339.67326.65326.5378823112343.161.11056275.71421.379423.62627858504490290.1644.0144.13160.5276.2474021.9616.34161233456.81400.645.633995268154273560620408832122948838301098094009314615621295048.994.2922.272593.28.829.70022553987028.018135.6875798.34986182.23572298.4094826284.802.98306335.1523.60479714712048.9919.603551.017248.470020.87231.17561849.5594.41187.87548.89204.21163.75363.3512.752421.30173767.4763.34966298.673236.8184.304.80418207.934290.279.7776449.386220.24955.05427197.8521034611217341032.413.71303591455.5831018.37.520.5661956.93305217.8290.72127.89156.133.17360315.03025.911525.57100.4613.54242282.2529623.778429.13165.785544566303560.89361631.05642.4422.585444.274862.64637.769.262120.3335.604928.0886883.0091036.5415.8947.9911.8638150.3918.081100.1859.935316.686323.251698.0561.328.584502.745.54379180.2256.565913.0456.580217.674712176.1757955.7381235.67175176.29227812558495349.2031230.0801218453.34287313537.7124.63101665.4690.5052.2271770.019.1117348.4291883452437321073627667136083334479.282.1596852.1422957.343.4513.10081.3554.8131.331912.5624806653.38540896672534.5503253333133.18810.9331995000282.5014691666675392266675426066675693915546666675.43847.643756939161046666718235666671797400000555.597156176666730.1761.2259.1579.832398690.0915.5125.7894515.09112.331.2574.95658439.11791455.2380.8646.3247.2601181.49429.7634.28911.8521342.0537567.6197398557.1615.50382266.3223.65823.970828410.71218.84424.1751.811.8122655.329.1044.87455.42082.8243.6733126.4407134.0728.20113.4945575.90657317.66484.821.1624.0936.6733.3931891.422134431.344390693.1681429.944.9502137.506193921167.128164223.5414185.116906.27.020.2893480909120.113214376.398237.0181.99539912422361.461249472887.077.08301740526.36528.21513.92246.0810677837949.556.38343161.93662.000114.912276267766403464.7855.1155.01182.2685.4971127.4318.1595527267.52540.9622.581990352059967171406176108710866181919509145597120076765271954716.8210.618.3531157.833.0855.25761985166422.788644.063593.568115135.516389.2758683410.797.09682141.83020.752593330757417.2842.887923.496090.062111.27072.24645446.8055.2496.16116.9779.4291.9661165.930.8576853.21781310.5913.80792263.52443.771095.796.17190161.947391.4055.3382137.9897.248698.39969119.558517855208931392.516.810595465331321.48.420.35178488.035613.9286.0843.171110.499.95245100.460022.208231.7263.4043.61792276.207121976.912348030.69489161051422.8366011.954013.9666.435515.051633.185766.624.0811735.4241.569624.05401897.64644091.0233.63205.0360.374823.844.919752.89114.1588.759713.7812665.6219.849.2520550.939.06607110.1685.3435809.59158.2656.3182817924.13429725.87240811.458087.25795831031209151309.2197730.4718448083.72998025977.418.670.5635.3166999.714.2954678.485111646574136644672813331308313322166.6019.40521560.5659466.622.5212.13022.52511.910745.26129400012.83554156673125.4849340000161.33943.3936040000248.0327728000001394733333239433333337626847593666678.513210.4367374589801270000079352666674407900000783.200227930000055.0511.4710.16738.3163.483.08110862.82112.274.0322.70798926.82362787.8241.647.75236.6105915.25331.9517.8392.07683519.92390520.93818551955.55116.122612296.2716.4013.753199018.99211.7204.1333.473.4588487.916.9554.510231.42723.62214.0178508.0926163.7955.95078.4961584.943244738.81664.753.7712.8901.1991.1094461.0019147885.162396176.4221450.527.5824121.963122574309.847653245.8214955.614143.114.080.2513989358123.371613828.748996.8290.34121452465952.9314476044813.9513.89300833595.19597.73566.31209.6812616102877.323.41085176.87646.891147.837138409588983425.2555.3855.56178.0325.6650827.6112.58079492123.90542.1516.219960071092678597205999635449040847734586513191338392565207734.6912.4315.8615786.624.6932.52321415841921.578446.8764178.25382118.54229124.4558634258.394.10914243.3409.335341959577734.7124.109741.474571.556413.97451.48596672.9168.48103.35234.9094.7697.004801.9611.251791.69307590.7073.57543279.641110.91433.066.16668162.092406.5832.5188102.7349.736858.13326122.94838645443491385.317.25376330347.5471324.98.450.45138605.5210378.96104.5283.11576.985.59491178.69813.526963.3161.6292.23922446.30374389.064260.256.912166031953542.17122521.632212.4133.499429.849332.002986.337.856017.1934.812628.72501392.91319827.2471.98101.5134.644744.878.485608.0781.431912.28076.936846.7941.179.429997.896.18478161.4975.4117230.35102.0359.8000816530.00120132.8893009.18524108.868546523310091187.6844692.1064152864.87538029384.935.66206619.2920.5734.0445656.925.3827287.26171606557911223761490001338733311424.0510.32241301.2653167.820.4517.01623.26515.2226.302732.7467132234.20554553333114.2850586667162.47529.1836066000330.2628620000001368733333157040000019289847553333338.595133.4769192574581190000058624666674353533333783.422228746666762.7018.6518.3640.918961868.5440.505.46066040.00112.278.1528.32559425.93164686.2981.2814.29151.9483798.69532.1219.9913.36045326.23225311.82818362511.4373.37966346.1516.6686.881212797.47112.1996.9502.662.6462071.106.8574.312155.12083.61130.7874500.7673165.7535.75745.2457944.770134994.14680.739.9982.7231.9841.3690631.1192310176.8642077238.8811575.528.3185140.013161216466.548014334.3012394.614217.113.950.2783603261138.111436324.935964.9311.82519770835053.7714210225414.0513.84298341559.55562.50536.82211.85125562060176.142.87877185.30042.507164.021126475137690470.9555.6155.59182.0175.5501827.6612.50279987139.25529.8014.298972884361237676965536675033110211233962757549735096099835734.6511.3415.5714207.721.3228.06891460854423.919842.0692325.42331136.39912135.6668220895.454.46135225.3088.182751802499335.0125.478939.258672.123213.86471.66046604.0458.35111.68834.90100.24108.02800.2311.251351.70579586.1785.34656187.012157.45306.277.88584126.816447.6128.0956101.46869.860188.15092122.68343290500191388.917.24285570547.4311328.58.380.48133779.6211809.01114.6391.16525.885.25321190.38216.325697.5161.9922.66711374.74546233.742172.462.173188962230546.50139127.461743.4934.163629.269034.922739.018.585528.1533.520929.83811363.33294183.2510.1093.9930.607348.889.475024.3281.604112.25497.236554.1240.4510.099356.716.37780156.6105.8516038.90102.1919.7851616034.86617334.5672649.08955109.998511409272949171.7714294.2833926265.27712027695.240.26186675.4320.5734.1485239.625.7526082.71164262654950022828020001339466710652.039.18441278.5850909.622.5316.73024.23712.5145.630131.4464347532.80554253333105.9811373333172.87528.0136068333354.5928598333331158166667136216666716654142563000008.422113.2783166541526646666752719333334032733333783.778228276666762.0221.0220.1543.454051942.1939.585.91645838.90112.276.8332.59178328.03102287.9931.3115.96136.9453423.63431.7220.4943.75483532.56125191.11818269047.2865.30255229.4116.9638.253835116.74313.4717.7382.732.7353182.777.5283.850132.49183.62110.5253449.1141178.5165.79939.0862624.792116725.04684.736.1512.7562.0581.4229541.154876918.0213335145.6471558.442.5158161.906162228248.867315266.1012708.511941.37.140.2933422731133.872861959.871942.0197.63037075153061.071255956077.207.04299610525.23529.80504.84240.4310882158077.645.27839164.89957.099118.902251645629887525.8452.7952.81220.3514.5381726.4118.0595550579.05529.0317.891785006879067152650546581310015026348607882225738369892582222316.2410.568.2726702.729.6147.52672010828925.317139.8035124.29946140.4826194.8108498306.787.55558133.48517.811053032459617.0742.513723.578693.834010.66152.25020444.8475.24102.58317.4185.89102.981094.2560.9170433.17064315.3555.98436167.10763.98750.119.74614102.572401.8447.6143155.9606.4122910.434495.837220750279431382.117.08576871791324.68.300.35172688.506613.6682.9448.43990.169.96245100.358623.127810.6562.8873.68001271.62086757.781025.833.675104281242730.0578014.303353.7366.628515.008236.905178.294.4610708.9139.017625.62651862.79593246.9253.63188.7452.674126.355.209203.19113.5178.809303.9812024.1819.3410.1718666.169.12452109.4645.9831974.88157.5056.3489117825.84426227.67836811.502486.9230553815145110291.2787281.9587975907.45064922931.821.120.5735.4026835.714.4748760.815103327776173443249303331339066719414.9417.00981524.7443531.023.5312.63120.23410.601842.03125333312.42549546673098.5824583333167.22340.9636033000227.2227302333331199966667212953333332052742809333338.349175.5971316492741186666775133333334063333333782.322228206666753.4412.8711.25727.7761.113.40833761.07112.272.6825.00563227.73771286.3411.678.61217.8465446.16128.8118.0892.31297719.88256584.67822472931.04105.405210139.6916.4094.260785968.19612.7494.6093.493.4884477.727.6454.329185.71103.62176.2061491.5344171.2185.84668.8685054.829194942.32673.450.1212.8101.2421.1263291.00947311895.4072508360.207170023.6266135.173484477534.059428400.5212731.89839.6411.010.2923422492125.761872096.153966.9337.68518976379958.1113178130910.9110.88250695458.96450.12433.79227.40116685326231.962.88869212.54647.986186.103161269162280450.6147.3147.33166.7936.0116423.3613.84372244114.76425.4616.7911672211418709975390078736563963293875274663411345703974887027.5312.4113.059320.523.7635.14641577618626.445837.9585449.46597167.40625146.7617736018.746.22791160.9047.997562282942627.6832.474430.857287.655211.41822.61653382.1037.47128.27927.78106.54110.80319.3483.131562.21020452.3237.63337131.004230.24283.1811.139289.7350462.7935.2461120.5568.295339.42177106.13037383429451139.913.85899510706.4651102.26.680.73132973.439546.19106.14102.84621.426.52367153.24622.675310.0773.2713.12904319.45228854.328476.648.849152861812948.36113038.211671.2340.633024.609538.853280.2510.975778.5136.446027.44711620.53374413.5551.07115.7837.738139.8311.355594.81103.5979.652968.037862.6531.6210.9211508.197.28087137.1936.2519917.54115.7908.6359415133.12020738.52630410.553794.7365535573235452232.7495818.7364525288.00301425194.631.65177989.8910.4840.3515973.820.6424218.83178087157544022753526671117450012575.3012.31871454.7150139.923.6914.78521.76213.2427.557332.4480318721.00462700002596.1709650000163.41329.0029907333314.6724449333331369800000174470000022018847145000007.283133.9297220188640136666765882000003791566667655.795189876666754.4617.7016.5845.078087924.8039.085.14765838.47112.168.5633.81422330.38137278.7391.5012.86229.9135747.82328.1623.2173.65430335.08127519.37974279397.4586.01395702.1319.9068.872259146.92816.3886.3503.163.1243799.788.8013.968156.39093.04126.1456527.7997169.2306.85344.9708265.662125567.07596.240.3863.2511.9211.3557370.9436217365.8634012201.2651712.641.8727174.935861750253.318553349.6711208.713455.05.710.3293044620131.383729470.8107830.6210.19730120223266.541153178345.755.79251075427.95420.35411.26253.1310363048393.995.83606192.97758.839147.735320258686237588.4347.1347.14203.5714.9142023.2816.8845922865.58422.7622.0023166119452831983103026700130692850757310477663304790602860899013.769.656.6417873.132.2656.44852290015227.897535.8712137.11385192.41705101.5617645325.708.51808117.61718.974543659320413.6855.915818.013393.636010.751883.16441316.1134.63126.11213.1190.85111.15397.4292.517093.69602270.6317.84065127.53785.86750.9611.577486.6864397.4456.6936148.2086.7497710.2310698.637416960200711139.013.711699473851104.26.640.41166151.425312.4767.0856.731127.0511.314888.374128.106513.3674.6893.98735250.63454066.052639.526.77684131001527.7463015.844032.6683.766011.938239.866400.94512750.5938.000026.32571876.66759922.1275.53231.8363.845921.666.2710182.16125.2447.984024.3714601.5214.9011.2022499.0310.066699.21826.638546.66179.9895.5576716427.14531030.68040312.809378.06765671135206564288.5377213.4338171241.70851229877.116.820.4841.9168396.710.9446126.080134022677696827830390831127366722017.7522.00231403.4642113.527.8611.55721.30514.103838.1215510008.10462296672590.8709393333159.74637.1329893333176.0023747666671349800000173480000041272346265666677.153200.1612415412631626666764390000003794133333655.046189496666746.0011.219.71658.9754.572.94045654.88112.159.0124.57012229.47448477.5961.717.17214.6175365.42128.1121.2242.08438418.59214017.18972588329.32121.30179771.4119.5584.510618937.91515.8103.9813.573.5963265.018.8484.100208.31313.03190.0839462.2382168.7796.98684.6885945.740245447.72576.851.6293.3411.3591.0683980.82965316146.9971266468.6882024.722.1972101.17851997567.286947495.255826.447382.4313.260.384260650098.70881480.825894.6406.4149919814465.6611687979913.2813.19250024338.84337.32323.59251.47105432862297.251.51901242.93127.532277.52574584721007313.5046.8746.88216.7034.6649123.2516.32161272240.13421.999.395397849313734623097505303070689383772437846082021118566172733.767.3815.254660.013.8117.9448933364527.928136.0772691.64224163.69864205.0146198256.724.57029219.0585.706721205390133.4932.920130.374380.723212.38842.25716443.0585.16154.01033.68176.87137.23388.7432.572781.82443547.5247.87325127.085161.48197.9111.204789.2081308.4617.9668112.7288.874469.38681106.52672051842611118.813.92740917786.1181101.46.760.5387511.11204310.7987.31100.67317.384.82506207.33617.303636.3375.8643.91006255.67116573.216704.4105.676388114537757.38242131.99998.0936.611927.312841.381542.6110.782943.0835.794327.9362875.23172566.8592.2253.8918.470585.1211.222830.12102.05039.80098.593692.6741.1111.785366.797.74521128.9826.779256.11114.1748.7581412363.79611244.27617210.302697.0508379266186537107.7152692.8732274636.13154419375.967.67153818.0310.4841.6573193.021.0016949.5913679893313571150052667112900006416.345.7420843.6324526.033.2313.96845.9536.4993.514619.2538621236.66462106672593.3633860000160.78517.0429899667276.93218683333380991666782101333310222229828000006.93964.3253102198311693333330622000003047066667655.239187133333350.0435.8234.3556.417518673.4224.3111.3696423.71112.154.6058.16923533.67316268.1880.8627.12107.3532683.83128.1327.5366.54176834.5152745.64976133284.4641.02252746.2021.03520.937913910.20517.91713.2501.801.7922229.948.9295.07775.21493.0362.1866294.6209165.0897.18221.8175775.73563930.64557.924.2463.4843.8012.2004971.4983516609.7811177496.2412224.622.134195.63771352568.254049498.225821.467421.2313.180.414241637592.82810291.821047.5438.1689065815664.9311788497713.2813.15250162335.46337.44322.69251.05105138551300.691.46163251.60625.334288.11867806730387295.2946.8646.93226.6264.4653323.2718.05355395263.76418.898.484889653908574158939632072820533270772219447097319330509364333.186.9315.114564.313.0016.4505893417626.829037.5302690.73503188.37105215.9255820618.184.74221211.1875.449711188210333.2332.229431.064486.651511.54472.23868446.6895.11159.62433.30187.03151.27443.9612.254531.80573553.4717.86015127.208156.59204.0511.286488.5688287.4216.4504114.3118.752339.71131102.96678093916431144.813.82429401106.0001096.56.740.678334.92224010.7080.86109.00293.074.92653203.05518.493405.9478.2413.92455254.72716399.715903.4120.560425134884457.54264934.76918.5338.102726.247545.711396.7811.482763.4835.883127.8703796.82153920.8654.8348.7316.920094.4112.672506.10112.7608.871449.463355.7340.9713.114830.017.70511129.6467.608279.63118.0238.4728312066.70510546.35216310.397796.167936323499263697.9782449.4542069432.52809319485.875.26141831.7760.4842.6052818.520.6516855.8913383553190941147108000112363335745.045.2406771.4724548.835.9513.61454.3896.2463.199317.8033999836.48454863332593.0577263333153.41715.7729899333256.7920263333337317566677417400009347226870666676.72763.343393473279370000027896000002767933333656.250177686666744.9238.8638.0461.362463617.0323.0212.1513222.46112.151.5159.26637736.82344164.7870.8030.0798.2072455.16327.7428.4206.85271234.3552684.93974130839.1237.95972718.9421.53021.178030710.42718.17514.4931.691.6921684.808.9485.16573.90733.0358.2776273.6703157.3627.30120.9977325.73462449.55563.621.9353.6234.5452.3211951.797492020113.422239294226.606.110.8151229649103.221479879.022587.4192.098235445499107.28716640056.336.15197298320.87320.17311.63380.33675997930.70757252.968187.654256330349777354.8126.9026.89194.1075.1516613.9028.80734716106.48258.497.7110960821997139794193887973528268216537859939523024000513515314.816.416.4710431.616.8832.79792446614347.197721.1982138.5661783869.9612.434580.41992.554002763078114.84179.1445.58241155.9006.604713.61920276.2352.56160.46214.67608.0941.6467412.302181.25726.14327178.1131293.02245.277.32610136.391301.4632.9663133.3567.4994016.806259.49836244971583721.410.61158496546701.25.601.2075761.5014971.7661.48294.69243.7636.400927.4727340.26843.92143.79710.806692.508358.957382284390628.99183241.501731.50285.9883.49785180.291593.8816.284376.0956.773417.6172979.89456271.0719.8399.7221.222735.1126.122743.39149.9946.6671818.753787.9416.4528.2410020.9117.055358.576719.2314551.12217.9294.5896011833635.54635016.582860.2880474111572074278.2351955.8816575205.53864631677.227.410.3579.2003067.02.8342946.258767313591835628076711.3573857.2039580.929.3857.129122.033676575.88334393331760.8349746667108.29320.4631705667168.101852366667707053333130673333336313337121333333.710125.1146363300680090000068215333333607900000494.379178053333327.3318.7016.52464.8029.1529.1549.836.4042.8571.5110.5676.6021915.04122.0725.71116.24111971.024725.8797.96720.5886.0452.972.9812.2453.629138.84992.23121.4320338.1468108.4959.96031.4785578.094293.14.8292.6051.8576422.49734126791.961508453314.8811.880.572174865887.17743282.711348.5274.85512279424690.318511789711.7411.95196740379.92381.05365.62312.32819004770.52357267.951246.848129975994733278.0527.1727.17188.3005.3105014.0112.62879191191.20261.199.405657207677835063013746231772490621834060513792312059178725726.917.9013.395315.715.9019.42511134038941.083324.3394184.2022299150.517.68977130.0623.768111754893426.89106.6889.37294124.0648.060132.24323445.6235.39165.40926.82562.3981.778095.66426176.6434.80725207.942283.83126.647.16440139.454260.8319.515978.730212.701516.863159.298988579101444743.510.65969554944.052739.15.651.3684584.1323184.6771.41259.66138.5317.900055.862282.121747.50140.8594.20355237.803115.462520835765738.01273138.87924.90160.1286.24521172.72832.2915.112366.8342.901623.3078677.40233836.3692.7951.8512.173765.6725.331414.86116.6448.5728617.812006.0032.4527.015278.467.59589131.53618.207721.00140.1447.1358714525341.09527014.476169.068350457474955239.348983.7072915609.74736827281.952.47114715.7890.3576.4741953.18.5626024.60242106341385162843005.8632609.6740441.010.39157.7133.650515.6518892512.40334446671765.6359576667114.54213.7731726000214.14190626666771962666784837666719005837363000003.97276.5292190209425943333343011666673671966667494.046184140000040.7329.0629.5558.306715443.7419.3418.8649.841.8646.4830.9819.8539.171979.27920.2227.85029.56111740.692425.88511.19820.90811.4791.931.9310.9965.17784.70572.2373.9558284.4616115.3459.18815.9416047.507299.44.2434.8062.5046212.29712139197.671640921313.5013.640.542184608092.42826049.512610.8251.19512519406684.599097341413.6113.00220051400.52403.96388.29296.36871903970.56717238.432228.571144163027280298.1629.9429.95174.1405.7423415.4711.53786677174.49284.3310.096384532037635689375783131969206832704508875564013403303909030.878.4714.535753.915.8321.51471186481536.783427.1844169.4472339570.286.97509143.3553.763481794574530.6996.687010.3425111.8748.938332.07288482.2095.58152.22930.86515.0791.941445.24051190.7824.59703217.445266.91134.776.74630148.107281.6221.622272.312013.829515.105766.19718100392175816.711.56731592174.440797.76.191.2591273.0220905.0276.37240.38149.6315.585664.158475.221906.99126.9664.06272246.037103.863468755250138.01247235.301018.18146.7796.81330153.76934.9714.012553.5538.419726.0268732.79262432.6618.9758.0613.401860.1123.491525.26105.2239.5035316.362179.9935.8524.255866.637.13535140.02616.418536.39129.9407.6958315528137.64029913.178975.865056664530948343.7281093.2063340274.03430929804.448.07116716.4010.3969.3432118.69.6026864.11046376843804869839336.5166668.3340626.411.18152.0744.057417.2920980313.14371956671960.7398566667124.59715.2435259000232.51209903333379634000094027000021131141367666674.35083.7321211449471470000047991333334102200000540.240205730000044.4426.4926.5252.321381482.0621.3920.8752.546.1850.8221.0717.6143.5531088.83321.3025.23531.99100745.210023.40510.38619.20310.3012.112.1110.2434.77092.87842.4981.8340314.6232126.5858.35917.8536826.849333.23.8724.3242.2799712.049102228124.602449334236.816.610.7751292122115.121644363.725098.0174.272266919825101.49757401396.656.59217536337.75339.50333.06377.58683208000.57694233.205173.087287035041140378.7928.3928.39182.3615.4834314.8726.2883804397.18283.527.99126688286665311253730411733919112254808919441685726658385788715.976.747.1011147.217.4536.24642705978643.806822.8289125.2741821144.0011.482087.08672.397702765777616.10167.9745.95283141.5747.312853.41537292.7912.67148.99316.22587.0001.7036512.126882.44214.74859210.365274.52261.6410.58571103.8947335.3836.5043122.7518.1461515.127466.17725608266487774.411.61324163340753.26.101.0976895.2913421.9968.52271.14264.6635.226628.3969293.36978.46129.91010.588394.424753.015350813928328.92163137.171933.08266.1043.76003157.281826.8214.304992.7854.254318.42951064.18519392.2630.23113.6823.116331.2824.242957.69140.077.1391716.734259.9917.9124.7711390.0416.541860.388316.9816394.06219.1734.5627812638132.65538915.278465.4320517127290756186.8842172.1037302247.13820633010.524.610.3871.9683414.22.9944136.525846335611200696996712.62201010.9337461.126.4627.906022.594072566.72371213331954.2384406667116.61821.6335190667190.802036733333777940000143710000040227540686666674.040136.2218404739759756666775463666674021033333538.908198930000028.4316.8915.16520.0131.7031.8552.638.9247.2531.689.7184.3472108.67519.5523.32318.34100579.285223.4567.46818.8515.5123.313.3111.6193.411151.24812.47133.4150363.3844117.6819.35135.1966927.603323.74.4862.3371.7180222.24526OpenBenchmarking.org

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterWRF 4.2.2Input: conus 2.5kmXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4K8K12K16K20K5601.1010602.416674.1211980.8114030.5118998.826840.0612410.297157.9613273.697365.8611895.414431.347885.166918.0210176.8616609.7816147.001. (F9X) gfortran options: -O2 -ftree-vectorize -funroll-loops -ffree-form -fconvert=big-endian -frecord-marker=4 -fallow-invalid-boz -lesmf_time -lwrfio_nf -lnetcdff -lnetcdf -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 2.0.0Benchmark: vklBenchmarkCPU ISPCXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P9001800270036004500SE +/- 13.62, N = 3SE +/- 2.08, N = 3SE +/- 6.74, N = 3SE +/- 1.53, N = 3SE +/- 12.02, N = 3SE +/- 6.81, N = 3SE +/- 4.36, N = 3SE +/- 13.96, N = 3SE +/- 8.62, N = 3SE +/- 6.66, N = 3SE +/- 3.61, N = 3SE +/- 0.67, N = 3SE +/- 7.31, N = 3SE +/- 0.58, N = 3SE +/- 26.33, N = 3SE +/- 0.88, N = 3SE +/- 5.36, N = 3SE +/- 1.00, N = 3SE +/- 8.67, N = 3SE +/- 1.20, N = 3SE +/- 8.17, N = 3SE +/- 1.20, N = 3322220042716173415278691852135617761243222813912020126740122508390623963335207711771266

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 4.0.1Test: Basic - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P150300450600750SE +/- 1.41, N = 3SE +/- 2.68, N = 7SE +/- 2.65, N = 3SE +/- 7.14, N = 9SE +/- 0.35, N = 3SE +/- 7.86, N = 9SE +/- 2.66, N = 12SE +/- 4.65, N = 9SE +/- 3.24, N = 12SE +/- 2.98, N = 3SE +/- 4.04, N = 12SE +/- 7.69, N = 9SE +/- 1.27, N = 15SE +/- 2.17, N = 3SE +/- 2.11, N = 12SE +/- 2.95, N = 3SE +/- 6.20, N = 3SE +/- 4.63, N = 3155.74290.45203.41373.18368.22705.58201.38384.38214.42412.37201.27360.2193.17176.42145.65238.88496.24468.691. (CXX) g++ options: -fopenmp -std=c++11 -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -ljpeg -lmpi_cxx -lmpi

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P80016002400320040001809.71878.31960.92070.32170.03576.41873.12245.21785.02530.51712.61700.01429.91450.51558.41575.52224.62024.71. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz -lcomex -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2

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.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 144 144 144 - RT: 60Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1530456075SE +/- 0.94, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.34, N = 9SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.37, N = 3SE +/- 0.29, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.51, N = 3SE +/- 0.28, N = 5SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 369.3035.1460.8531.0730.1120.4162.6430.5954.4125.2441.8723.6344.9527.5842.5228.3222.1322.201. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 512 - Model: ResNet-50Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.61, N = 3SE +/- 0.48, N = 3SE +/- 1.39, N = 3SE +/- 0.01, N = 3SE +/- 0.21, N = 3SE +/- 0.08, N = 3SE +/- 1.80, N = 3SE +/- 0.61, N = 3SE +/- 0.79, N = 3SE +/- 1.17, N = 3SE +/- 1.22, N = 3SE +/- 0.04, N = 3SE +/- 0.93, N = 9SE +/- 0.01, N = 3SE +/- 0.81, N = 3SE +/- 1.62, N = 4SE +/- 0.31, N = 3SE +/- 1.42, N = 4SE +/- 1.83, N = 3SE +/- 0.34, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3169.75140.53174.95148.45127.9791.75138.39125.14145.29117.53124.6097.67113.4291.96174.93135.17137.50121.96161.90140.0195.63101.17

BRL-CAD

BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.38.2VGR Performance MetricXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1.3M2.6M3.9M5.2M6.5M5186677916904413696983389223009474308854075342767812365879071393724493341640921223929415084535861750348447761939213122574616222831612167713528519971. (CXX) g++ options: -std=c++17 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lnetpbm -lregex_brl -lz_brl -lassimp -ldl -lm -ltk8.6

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: X3D-benchmarking input.i3dXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P120240360480600SE +/- 0.37, N = 3SE +/- 2.51, N = 3SE +/- 0.05, N = 3SE +/- 3.88, N = 4SE +/- 1.19, N = 3SE +/- 0.13, N = 3SE +/- 0.01, N = 3SE +/- 0.14, N = 3SE +/- 0.04, N = 3SE +/- 1.25, N = 3SE +/- 9.13, N = 9SE +/- 5.39, N = 3SE +/- 2.20, N = 12SE +/- 3.06, N = 9SE +/- 3.26, N = 9SE +/- 4.73, N = 3SE +/- 0.22, N = 3SE +/- 0.92, N = 3171.38323.25196.81357.01292.03558.38187.22351.64189.69407.59253.32534.06167.13309.85248.87466.55568.25567.291. (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

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeaf 1.3Input: clover_bm16Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P110220330440550SE +/- 2.02, N = 3SE +/- 0.54, N = 3SE +/- 0.74, N = 3SE +/- 0.13, N = 3SE +/- 3.41, N = 3SE +/- 1.58, N = 3SE +/- 2.11, N = 9SE +/- 2.03, N = 3SE +/- 1.56, N = 3SE +/- 1.43, N = 3SE +/- 2.04, N = 3SE +/- 0.00, N = 3SE +/- 1.60, N = 12SE +/- 0.11, N = 3SE +/- 1.02, N = 3SE +/- 4.99, N = 9SE +/- 0.59, N = 3SE +/- 2.24, N = 3SE +/- 0.12, N = 3SE +/- 3.79, N = 9SE +/- 1.08, N = 3SE +/- 0.79, N = 3378.05410.59214.80328.47295.30512.36269.20357.57263.33351.29236.81313.50226.60314.88349.67400.52223.54245.82266.10334.30498.22495.251. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

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.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - GriddingXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3K6K9K12K15KSE +/- 60.72, N = 3SE +/- 54.27, N = 5SE +/- 75.19, N = 5SE +/- 4.15, N = 3SE +/- 19.36, N = 3SE +/- 6.18, N = 3SE +/- 195.28, N = 8SE +/- 32.87, N = 12SE +/- 23.56, N = 3SE +/- 87.89, N = 12SE +/- 21.01, N = 3SE +/- 2.09, N = 3SE +/- 204.28, N = 3SE +/- 262.30, N = 12SE +/- 141.43, N = 5SE +/- 59.24, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 36038.925062.077630.068088.115279.075172.964382.973406.963347.933375.3211208.7012731.8014185.1014955.6012708.5012394.605821.465826.441. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - DegriddingXeon Platinum 8380 2PXeon Platinum 8380EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4K8K12K16K20KSE +/- 322.56, N = 3SE +/- 3.76, N = 3SE +/- 345.37, N = 3SE +/- 25.80, N = 3SE +/- 889.21, N = 3SE +/- 212.00, N = 12SE +/- 267.65, N = 5SE +/- 128.32, N = 3SE +/- 0.54, N = 3SE +/- 34.17, N = 39893.327425.2213455.009839.6416906.2014143.1011941.3014217.107421.237382.431. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-152Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.16, N = 3SE +/- 0.29, N = 3SE +/- 0.13, N = 7SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.18, N = 4SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.15, N = 3SE +/- 0.04, N = 3SE +/- 0.19, N = 3SE +/- 0.06, N = 9SE +/- 0.04, N = 3SE +/- 0.12, N = 9SE +/- 0.15, N = 3SE +/- 0.13, N = 9SE +/- 0.20, N = 3SE +/- 0.13, N = 9SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.06, N = 316.0023.3314.0820.2514.8817.7014.9219.5816.3019.206.6113.646.1111.885.7111.017.0214.087.1413.9513.1813.26

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average LatencyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.29090.58180.87271.16361.4545SE +/- 0.040, N = 12SE +/- 0.005, N = 12SE +/- 0.030, N = 12SE +/- 0.007, N = 3SE +/- 0.022, N = 9SE +/- 0.005, N = 3SE +/- 0.038, N = 9SE +/- 0.007, N = 3SE +/- 0.017, N = 9SE +/- 0.008, N = 12SE +/- 0.008, N = 12SE +/- 0.006, N = 5SE +/- 0.009, N = 12SE +/- 0.007, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.006, N = 12SE +/- 0.002, N = 3SE +/- 0.004, N = 12SE +/- 0.003, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 31.1490.5891.2930.5720.5250.5811.0980.6221.0270.5590.7750.5420.8150.5720.3290.2920.2890.2510.2930.2780.4140.3841. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P900K1800K2700K3600K4500KSE +/- 30933.90, N = 12SE +/- 13863.97, N = 12SE +/- 18635.21, N = 12SE +/- 20350.34, N = 3SE +/- 74539.48, N = 9SE +/- 13469.59, N = 3SE +/- 36627.33, N = 9SE +/- 17697.00, N = 3SE +/- 16726.04, N = 9SE +/- 24978.23, N = 12SE +/- 13233.94, N = 12SE +/- 19583.16, N = 5SE +/- 13874.42, N = 12SE +/- 20221.52, N = 3SE +/- 13473.87, N = 3SE +/- 15503.17, N = 3SE +/- 71870.42, N = 12SE +/- 27594.68, N = 3SE +/- 41074.86, N = 12SE +/- 39970.22, N = 3SE +/- 4774.62, N = 3SE +/- 11512.67, N = 38818811699046778141174814619300061720762920725160742497603517926361292122184608012296491748658304462034224923480909398935834227313603261241637526065001. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 256 - Model: ResNet-50Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.21, N = 3SE +/- 0.58, N = 3SE +/- 1.57, N = 3SE +/- 0.03, N = 3SE +/- 0.46, N = 3SE +/- 0.02, N = 3SE +/- 1.04, N = 12SE +/- 0.63, N = 3SE +/- 1.40, N = 3SE +/- 0.14, N = 3SE +/- 0.54, N = 3SE +/- 0.04, N = 3SE +/- 0.48, N = 3SE +/- 0.01, N = 3SE +/- 0.18, N = 3SE +/- 2.00, N = 12SE +/- 0.50, N = 3SE +/- 1.39, N = 4SE +/- 0.57, N = 3SE +/- 0.66, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3150.33129.02161.06143.67125.9191.65135.28123.69141.95119.73115.1292.42103.2287.17131.38125.76120.11123.37133.87138.1192.8298.70

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.orgverify/s, More Is BetterOpenSSL 3.3Algorithm: RSA4096Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P800K1600K2400K3200K4000KSE +/- 6745.42, N = 3SE +/- 92.62, N = 3SE +/- 862.96, N = 3SE +/- 92.46, N = 3SE +/- 40228.60, N = 3SE +/- 198.68, N = 3SE +/- 277.84, N = 3SE +/- 200.66, N = 3SE +/- 355.69, N = 3SE +/- 165.80, N = 3SE +/- 205.33, N = 3SE +/- 23.84, N = 3SE +/- 37.76, N = 3SE +/- 9.55, N = 3SE +/- 2222.55, N = 3SE +/- 444.74, N = 3SE +/- 6125.37, N = 3SE +/- 1576.96, N = 3SE +/- 5.26, N = 3SE +/- 149.48, N = 3SE +/- 136.07, N = 3SE +/- 126.84, N = 32057021.61036735.61916162.9974189.91099481.5587354.81627941.4813667.61395446.4697707.21644363.7826049.51479879.0743282.73729470.81872096.13214376.31613828.72861959.81436324.9810291.8881480.81. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.3Algorithm: RSA4096Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20K40K60K80K100KSE +/- 56.68, N = 3SE +/- 15.15, N = 3SE +/- 14.34, N = 3SE +/- 9.84, N = 3SE +/- 108.70, N = 3SE +/- 40.37, N = 3SE +/- 83.09, N = 3SE +/- 20.13, N = 3SE +/- 6.92, N = 3SE +/- 19.45, N = 3SE +/- 8.20, N = 3SE +/- 1.63, N = 3SE +/- 8.25, N = 3SE +/- 1.37, N = 3SE +/- 80.37, N = 3SE +/- 27.42, N = 3SE +/- 247.60, N = 3SE +/- 98.71, N = 3SE +/- 81.38, N = 3SE +/- 29.11, N = 3SE +/- 12.56, N = 3SE +/- 22.17, N = 361499.631594.454601.127431.136371.618389.348534.825441.545531.022861.025098.012610.822587.411348.5107830.653966.998237.048996.871942.035964.921047.525894.61. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: allmodconfigXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P130260390520650SE +/- 0.56, N = 3SE +/- 0.72, N = 3SE +/- 0.86, N = 3SE +/- 0.80, N = 3SE +/- 0.92, N = 3SE +/- 0.35, N = 3SE +/- 0.73, N = 3SE +/- 0.67, N = 3SE +/- 0.56, N = 3SE +/- 0.61, N = 3SE +/- 0.90, N = 3SE +/- 0.67, N = 3SE +/- 0.81, N = 3SE +/- 0.76, N = 3SE +/- 1.31, N = 3SE +/- 1.05, N = 3SE +/- 0.98, N = 3SE +/- 0.37, N = 3SE +/- 0.49, N = 3SE +/- 0.49, N = 3SE +/- 0.78, N = 3SE +/- 0.91, N = 3189.74316.90218.57362.45356.99612.78231.03384.06246.94428.39174.27251.20192.10274.86210.20337.69182.00290.34197.63311.83438.17406.41

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 1024 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess BenchmarkXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P90M180M270M360M450MSE +/- 2482737.87, N = 3SE +/- 2966712.90, N = 9SE +/- 2148606.38, N = 4SE +/- 1304590.97, N = 3SE +/- 1325249.25, N = 3SE +/- 1466028.98, N = 9SE +/- 2925715.37, N = 11SE +/- 629927.17, N = 3SE +/- 4344961.90, N = 10SE +/- 1751369.24, N = 12SE +/- 3095674.84, N = 3SE +/- 1441726.26, N = 3SE +/- 2692064.58, N = 4SE +/- 3565305.92, N = 15SE +/- 7961094.38, N = 12SE +/- 4652276.19, N = 12SE +/- 919073.22, N = 3SE +/- 2268830.04, N = 3SE +/- 2356843.91, N = 3SE +/- 3170744.86, N = 15SE +/- 2147688.02, N = 12SE +/- 2397455.91, N = 122226103781097663451877617699688747910459720051099545149409082735922301379772036801908826691982512519406623544549912279424630120223218976379939912422321452465937075153019770835090658156991981441. (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 -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto-partition=one -flto=jobserver

Apache IoTDB

Apache IotDB is a time series database and this benchmark is facilitated using the IoT Benchmaark [https://github.com/thulab/iot-benchmark/]. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20406080100SE +/- 0.80, N = 7SE +/- 0.68, N = 3SE +/- 1.18, N = 3SE +/- 0.15, N = 3SE +/- 0.53, N = 3SE +/- 0.43, N = 3SE +/- 1.11, N = 4SE +/- 0.87, N = 3SE +/- 1.01, N = 3SE +/- 0.48, N = 3SE +/- 0.65, N = 3SE +/- 0.15, N = 3SE +/- 1.37, N = 3SE +/- 0.11, N = 3SE +/- 0.79, N = 12SE +/- 0.10, N = 3SE +/- 0.59, N = 12SE +/- 0.23, N = 3SE +/- 0.65, N = 12SE +/- 0.31, N = 3SE +/- 0.03, N = 3SE +/- 0.31, N = 382.7381.7392.2385.16102.67100.2995.3989.5394.8490.02101.4984.59107.2890.3166.5458.1161.4652.9361.0753.7764.9365.66

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P30M60M90M120M150MSE +/- 862582.24, N = 7SE +/- 739343.51, N = 3SE +/- 928557.10, N = 3SE +/- 276414.35, N = 3SE +/- 365828.34, N = 3SE +/- 306106.55, N = 3SE +/- 872659.60, N = 4SE +/- 782176.98, N = 3SE +/- 594234.55, N = 3SE +/- 435900.66, N = 3SE +/- 207721.96, N = 3SE +/- 174748.00, N = 3SE +/- 769517.20, N = 3SE +/- 84917.07, N = 3SE +/- 1225528.40, N = 12SE +/- 263772.09, N = 3SE +/- 1058559.19, N = 12SE +/- 538692.55, N = 3SE +/- 1249999.63, N = 12SE +/- 665173.47, N = 3SE +/- 41021.96, N = 3SE +/- 576868.09, N = 39277163994616807836517859087177075178480767294958013956686354518807704658574454275740139909734147166400585117897115317834131781309124947288144760448125595607142102254117884977116879799

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.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-152Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.18, N = 3SE +/- 0.16, N = 3SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.17, N = 3SE +/- 0.13, N = 3SE +/- 0.16, N = 3SE +/- 0.18, N = 7SE +/- 0.09, N = 3SE +/- 0.18, N = 7SE +/- 0.07, N = 3SE +/- 0.19, N = 3SE +/- 0.07, N = 3SE +/- 0.14, N = 3SE +/- 0.11, N = 9SE +/- 0.02, N = 3SE +/- 0.13, N = 9SE +/- 0.12, N = 3SE +/- 0.14, N = 9SE +/- 0.18, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 316.1723.4114.5920.4614.7617.5514.7419.2916.6819.476.6513.616.3311.745.7510.917.0713.957.2014.0513.2813.28

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-152Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.15, N = 12SE +/- 0.34, N = 3SE +/- 0.17, N = 3SE +/- 0.18, N = 3SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.17, N = 12SE +/- 0.21, N = 3SE +/- 0.19, N = 5SE +/- 0.05, N = 3SE +/- 0.13, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.15, N = 9SE +/- 0.10, N = 3SE +/- 0.13, N = 9SE +/- 0.17, N = 3SE +/- 0.03, N = 3SE +/- 0.16, N = 316.2523.4914.0320.1315.1117.7314.9619.2415.8419.016.5913.006.1511.955.7910.887.0813.897.0413.8413.1513.19

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.

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P70K140K210K280K350KSE +/- 411.11, N = 3SE +/- 209.97, N = 3SE +/- 330.95, N = 3SE +/- 625.56, N = 3SE +/- 285.52, N = 3SE +/- 372.88, N = 3SE +/- 196.24, N = 3SE +/- 670.43, N = 3SE +/- 384.08, N = 3SE +/- 256.85, N = 3SE +/- 1219.44, N = 3SE +/- 406.59, N = 3SE +/- 722.69, N = 3SE +/- 553.57, N = 3SE +/- 553.62, N = 3SE +/- 457.42, N = 3SE +/- 809.90, N = 3SE +/- 415.71, N = 3SE +/- 389.81, N = 3SE +/- 361.20, N = 3SE +/- 542.88, N = 3SE +/- 372.96, N = 33084343098962754722772052310762313572772982772602775892776472175362200511972981967402510752506953017403008332996102983412501622500241. (CC) gcc options: -pedantic -O3

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.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P130260390520650SE +/- 5.39, N = 5SE +/- 3.62, N = 6SE +/- 2.55, N = 3SE +/- 3.45, N = 3SE +/- 1.27, N = 3SE +/- 4.79, N = 3SE +/- 3.54, N = 12SE +/- 3.60, N = 3SE +/- 3.08, N = 3SE +/- 6.03, N = 3SE +/- 3.52, N = 3SE +/- 1.22, N = 3SE +/- 1.59, N = 3SE +/- 2.14, N = 3SE +/- 6.21, N = 5SE +/- 3.32, N = 3SE +/- 2.03, N = 3SE +/- 4.34, N = 3SE +/- 10.00, N = 3SE +/- 4.03, N = 3SE +/- 5.23, N = 3SE +/- 1.77, N = 3438.60458.23501.63481.66295.06342.99471.79476.91501.07470.02337.75400.52320.87379.92427.95458.96526.36595.19525.23559.55335.46338.84

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P130260390520650SE +/- 3.10, N = 5SE +/- 2.86, N = 6SE +/- 3.26, N = 3SE +/- 2.93, N = 3SE +/- 3.61, N = 3SE +/- 0.73, N = 3SE +/- 3.17, N = 12SE +/- 4.95, N = 3SE +/- 0.19, N = 3SE +/- 1.94, N = 3SE +/- 3.25, N = 3SE +/- 2.12, N = 3SE +/- 1.53, N = 3SE +/- 3.34, N = 3SE +/- 4.85, N = 5SE +/- 9.26, N = 3SE +/- 6.78, N = 3SE +/- 0.42, N = 3SE +/- 8.37, N = 3SE +/- 3.14, N = 3SE +/- 5.72, N = 3SE +/- 2.99, N = 3443.58455.79507.56486.05295.43339.67467.96474.68499.18465.06339.50403.96320.17381.05420.35450.12528.21597.73529.80562.50337.44337.32

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P120240360480600SE +/- 4.12, N = 5SE +/- 4.18, N = 6SE +/- 4.37, N = 3SE +/- 3.41, N = 3SE +/- 1.56, N = 3SE +/- 1.56, N = 3SE +/- 3.85, N = 12SE +/- 5.88, N = 3SE +/- 1.35, N = 3SE +/- 2.85, N = 3SE +/- 3.71, N = 3SE +/- 1.12, N = 3SE +/- 1.51, N = 3SE +/- 2.21, N = 3SE +/- 4.49, N = 5SE +/- 2.78, N = 3SE +/- 4.34, N = 3SE +/- 7.88, N = 3SE +/- 6.05, N = 3SE +/- 5.33, N = 3SE +/- 1.88, N = 3SE +/- 1.57, N = 3423.30437.20485.63465.50285.17326.65460.66461.62491.89446.40333.06388.29311.63365.62411.26433.79513.92566.31504.84536.82322.69323.59

Apache IoTDB

Apache IotDB is a time series database and this benchmark is facilitated using the IoT Benchmaark [https://github.com/thulab/iot-benchmark/]. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P80160240320400SE +/- 3.40, N = 12SE +/- 3.41, N = 3SE +/- 3.79, N = 8SE +/- 2.20, N = 3SE +/- 3.47, N = 3SE +/- 4.06, N = 3SE +/- 3.74, N = 12SE +/- 5.07, N = 3SE +/- 2.59, N = 12SE +/- 5.15, N = 3SE +/- 5.20, N = 3SE +/- 4.45, N = 3SE +/- 3.93, N = 12SE +/- 1.12, N = 3SE +/- 2.16, N = 15SE +/- 1.41, N = 3SE +/- 1.87, N = 3SE +/- 1.71, N = 3SE +/- 3.49, N = 3SE +/- 1.96, N = 3SE +/- 0.82, N = 3SE +/- 0.38, N = 3286.78280.26317.36294.33349.42326.53325.97309.03328.41314.44377.58296.36380.33312.32253.13227.40246.08209.68240.43211.85251.05251.47

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P30M60M90M120M150MSE +/- 761276.42, N = 12SE +/- 746822.47, N = 3SE +/- 692701.79, N = 8SE +/- 308756.34, N = 3SE +/- 661234.22, N = 3SE +/- 88241.50, N = 3SE +/- 574699.97, N = 12SE +/- 1037167.36, N = 3SE +/- 565297.90, N = 12SE +/- 662950.65, N = 3SE +/- 796440.21, N = 3SE +/- 441557.52, N = 3SE +/- 531367.14, N = 12SE +/- 111181.55, N = 3SE +/- 666882.79, N = 15SE +/- 68559.13, N = 3SE +/- 353097.82, N = 3SE +/- 371659.60, N = 3SE +/- 1152766.29, N = 3SE +/- 255887.16, N = 3SE +/- 412757.79, N = 3SE +/- 13050.78, N = 38923185092709751801696208779720674653007788231127805772883475645775012488245566468320800871903976759979381900477103630483116685326106778379126161028108821580125562060105138551105432862

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Chrysler Neon 1MXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P70140210280350SE +/- 0.14, N = 3SE +/- 0.20, N = 3SE +/- 0.11, N = 3SE +/- 0.15, N = 3SE +/- 0.78, N = 3SE +/- 0.13, N = 3SE +/- 0.14, N = 3SE +/- 0.18, N = 3SE +/- 0.13, N = 3SE +/- 0.08, N = 3SE +/- 1.00, N = 10SE +/- 1.11, N = 3SE +/- 0.46, N = 3SE +/- 0.49, N = 3SE +/- 0.71, N = 12SE +/- 1.51, N = 8SE +/- 0.50, N = 3SE +/- 0.47, N = 386.93157.56115.89207.20185.03343.16117.91212.99125.15228.7493.99231.9649.5577.3277.64176.14300.69297.25

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0b6Input: STMV with 1,066,628 AtomsXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P246810SE +/- 0.08185, N = 13SE +/- 0.02648, N = 15SE +/- 0.04403, N = 15SE +/- 0.01396, N = 15SE +/- 0.00911, N = 3SE +/- 0.00653, N = 3SE +/- 0.05463, N = 15SE +/- 0.05862, N = 15SE +/- 0.05644, N = 12SE +/- 0.05087, N = 15SE +/- 0.03216, N = 9SE +/- 0.00540, N = 3SE +/- 0.04439, N = 8SE +/- 0.00423, N = 3SE +/- 0.00332, N = 3SE +/- 0.00847, N = 3SE +/- 0.01739, N = 3SE +/- 0.00206, N = 3SE +/- 0.02927, N = 3SE +/- 0.00537, N = 3SE +/- 0.00540, N = 3SE +/- 0.00440, N = 32.254221.437871.109001.052741.612321.110561.767781.502141.886791.452230.576940.567170.707570.523575.836062.888696.383433.410855.278392.878771.461631.51901

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 23.0.1Time To CompileXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P60120180240300SE +/- 0.89, N = 3SE +/- 2.17, N = 3SE +/- 0.58, N = 3SE +/- 0.92, N = 3SE +/- 1.04, N = 3SE +/- 3.88, N = 3SE +/- 1.01, N = 3SE +/- 2.11, N = 6SE +/- 1.72, N = 3SE +/- 2.59, N = 4SE +/- 2.61, N = 3SE +/- 1.47, N = 3SE +/- 0.78, N = 3SE +/- 3.60, N = 3SE +/- 1.73, N = 3SE +/- 1.88, N = 3SE +/- 1.16, N = 3SE +/- 2.15, N = 3SE +/- 0.72, N = 3SE +/- 2.24, N = 4SE +/- 0.97, N = 3SE +/- 1.99, N = 9157.36187.51178.28210.93219.67275.71181.63215.50184.30220.57233.21238.43252.97267.95192.98212.55161.94176.88164.90185.30251.61242.93

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.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k AtomsXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.19, N = 3SE +/- 0.12, N = 3SE +/- 0.30, N = 3SE +/- 0.13, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 356.4037.4650.5234.8935.7021.3851.8232.1646.7128.2858.8447.9962.0046.8957.1042.5125.3327.531. (CXX) g++ options: -O3 -lm -ldl

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 21.7.2Time To CompileXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P90180270360450SE +/- 0.12, N = 3SE +/- 0.52, N = 3SE +/- 0.48, N = 3SE +/- 0.71, N = 3SE +/- 0.68, N = 3SE +/- 0.22, N = 3SE +/- 0.52, N = 3SE +/- 0.74, N = 3SE +/- 0.84, N = 3SE +/- 0.53, N = 3SE +/- 0.60, N = 3SE +/- 0.51, N = 3SE +/- 1.20, N = 3SE +/- 0.15, N = 3SE +/- 1.51, N = 3SE +/- 0.72, N = 3SE +/- 1.19, N = 3SE +/- 0.50, N = 3SE +/- 0.18, N = 3SE +/- 0.59, N = 3SE +/- 0.19, N = 3SE +/- 0.17, N = 3145.41214.43172.99254.00303.42423.63171.17252.65175.63270.38173.09228.57187.65246.85147.74186.10114.91147.84118.90164.02288.12277.53

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.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: SHA256Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P70000M140000M210000M280000M350000MSE +/- 29142783.86, N = 3SE +/- 23323764.84, N = 3SE +/- 66583540.43, N = 3SE +/- 159515825.53, N = 3SE +/- 919807437.76, N = 12SE +/- 8534071.27, N = 3SE +/- 70647143.49, N = 3SE +/- 33334799.97, N = 3SE +/- 126029257.64, N = 3SE +/- 27466135.54, N = 3SE +/- 228931567.43, N = 3SE +/- 242095363.59, N = 3SE +/- 97765007.32, N = 3SE +/- 34216643.88, N = 3SE +/- 80639802.67, N = 3SE +/- 34182873.86, N = 3SE +/- 160246527.97, N = 3SE +/- 146384349.91, N = 3SE +/- 210407867.05, N = 3SE +/- 38601729.93, N = 3SE +/- 64797308.82, N = 3SE +/- 10642001.84, N = 31319760501206657395792011910680210761806994523545307005762785850449010338286011351840436763886357534134443918518728703504114014416302728025633034977712997599473332025868623716126916228027626776640313840958898325164562988712647513769067806730387745847210071. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 512 - Model: GoogLeNetXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P130260390520650SE +/- 1.52, N = 3SE +/- 3.68, N = 3SE +/- 2.62, N = 3SE +/- 0.34, N = 3SE +/- 0.25, N = 3SE +/- 0.12, N = 3SE +/- 2.46, N = 3SE +/- 0.37, N = 3SE +/- 1.14, N = 3SE +/- 0.41, N = 3SE +/- 1.26, N = 3SE +/- 0.19, N = 3SE +/- 1.19, N = 3SE +/- 0.26, N = 3SE +/- 5.63, N = 6SE +/- 12.03, N = 12SE +/- 1.60, N = 3SE +/- 3.05, N = 12SE +/- 5.11, N = 3SE +/- 4.02, N = 10SE +/- 0.03, N = 3SE +/- 0.06, N = 3577.76462.78576.08461.23411.05290.16477.33398.40467.91383.57378.79298.16354.81278.05588.43450.61464.78425.25525.84470.95295.29313.50

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: PlatformXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1224364860SE +/- 0.15, N = 3SE +/- 0.14, N = 3SE +/- 0.12, N = 3SE +/- 0.16, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.27, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 353.2253.9248.7649.7943.5544.0146.5948.3545.9547.3228.3929.9426.9027.1747.1347.3155.1155.3852.7955.6146.8646.871. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: Video On DemandXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1224364860SE +/- 0.14, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.07, N = 3SE +/- 0.43, N = 3SE +/- 0.10, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 353.2953.9948.7749.8443.8044.1346.7948.1445.2947.3428.3929.9526.8927.1747.1447.3355.0155.5652.8155.5946.9346.881. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50100150200250SE +/- 0.44, N = 3SE +/- 0.71, N = 3SE +/- 0.69, N = 15SE +/- 0.09, N = 3SE +/- 1.55, N = 15SE +/- 2.32, N = 15SE +/- 2.59, N = 15SE +/- 6.41, N = 15SE +/- 5.82, N = 12SE +/- 3.36, N = 15SE +/- 0.13, N = 3SE +/- 0.21, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 2.91, N = 3SE +/- 2.29, N = 15SE +/- 2.18, N = 15SE +/- 4.76, N = 12SE +/- 0.35, N = 3SE +/- 4.84, N = 15SE +/- 7.19, N = 15SE +/- 6.87, N = 12100.74109.6792.35116.32110.97160.53121.78155.79121.67139.10182.36174.14194.11188.30203.57166.79182.27178.03220.35182.02226.63216.701. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.04271, N = 3SE +/- 0.05824, N = 3SE +/- 0.07619, N = 15SE +/- 0.00664, N = 3SE +/- 0.11577, N = 15SE +/- 0.08940, N = 15SE +/- 0.16487, N = 15SE +/- 0.22869, N = 15SE +/- 0.34632, N = 12SE +/- 0.15651, N = 15SE +/- 0.00381, N = 3SE +/- 0.00692, N = 3SE +/- 0.00154, N = 3SE +/- 0.00241, N = 3SE +/- 0.07023, N = 3SE +/- 0.08489, N = 15SE +/- 0.06453, N = 15SE +/- 0.16422, N = 12SE +/- 0.00718, N = 3SE +/- 0.15150, N = 15SE +/- 0.11914, N = 15SE +/- 0.14467, N = 129.926849.1188510.836238.597199.033676.247408.260166.549658.399627.241875.483435.742345.151665.310504.914206.011645.497115.665084.538175.550184.465334.664911. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: UploadXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P714212835SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 326.8026.9324.5624.9721.7321.9623.3023.7722.3423.3314.8715.4713.9014.0123.2823.3627.4327.6126.4127.6623.2723.251. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average LatencyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P714212835SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.12, N = 3SE +/- 0.12, N = 12SE +/- 0.12, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.15, N = 3SE +/- 0.03, N = 3SE +/- 0.14, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.32, N = 12SE +/- 0.01, N = 3SE +/- 0.25, N = 12SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 330.2013.2614.8614.0413.6916.3416.6613.8516.9614.2926.2911.5428.8112.6316.8813.8418.1612.5818.0612.5018.0516.321. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read WriteXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20K40K60K80K100KSE +/- 106.16, N = 3SE +/- 198.41, N = 3SE +/- 137.82, N = 3SE +/- 78.02, N = 3SE +/- 656.63, N = 3SE +/- 434.11, N = 12SE +/- 440.63, N = 3SE +/- 143.25, N = 3SE +/- 195.77, N = 3SE +/- 95.84, N = 3SE +/- 221.82, N = 3SE +/- 232.81, N = 3SE +/- 163.06, N = 3SE +/- 186.80, N = 3SE +/- 58.85, N = 3SE +/- 242.45, N = 3SE +/- 1059.49, N = 12SE +/- 87.23, N = 3SE +/- 865.71, N = 12SE +/- 232.97, N = 3SE +/- 299.71, N = 3SE +/- 175.33, N = 3331107539667315712277304261233600467222558952699693804386677347167919159228722445527279492555057998755395612721. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Barbershop - Compute: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P100200300400500SE +/- 0.21, N = 3SE +/- 0.69, N = 3SE +/- 0.42, N = 3SE +/- 0.18, N = 3SE +/- 0.25, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 1.09, N = 3SE +/- 0.23, N = 3SE +/- 0.12, N = 3SE +/- 0.17, N = 3SE +/- 0.08, N = 3SE +/- 0.41, N = 3SE +/- 0.33, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 3SE +/- 0.25, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3125.87236.73141.54261.97235.31456.81156.70290.56169.44320.1197.18174.49106.48191.2065.58114.7667.52123.9079.05139.25263.76240.13

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P120240360480600SE +/- 1.42, N = 3SE +/- 1.84, N = 3SE +/- 0.26, N = 3SE +/- 0.33, N = 3SE +/- 1.52, N = 3SE +/- 2.23, N = 3SE +/- 1.30, N = 3SE +/- 0.79, N = 3SE +/- 1.13, N = 3SE +/- 0.60, N = 3SE +/- 0.47, N = 3SE +/- 2.24, N = 3SE +/- 0.30, N = 3SE +/- 0.37, N = 3SE +/- 0.25, N = 3SE +/- 0.96, N = 3SE +/- 0.60, N = 3SE +/- 1.85, N = 3SE +/- 0.38, N = 3SE +/- 2.28, N = 3SE +/- 1.34, N = 3SE +/- 0.72, N = 3519.89520.22456.18455.57394.71400.64439.84464.72435.21461.61283.52284.33258.49261.19422.76425.46540.96542.15529.03529.80418.89421.99

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.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.24, N = 15SE +/- 0.02, N = 3SE +/- 0.23, N = 15SE +/- 0.06, N = 15SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.20, N = 15SE +/- 0.10, N = 3SE +/- 0.12, N = 3SE +/- 0.08, N = 3SE +/- 0.22, N = 15SE +/- 0.06, N = 3SE +/- 0.22, N = 12SE +/- 0.02, N = 3SE +/- 1.02, N = 15SE +/- 0.28, N = 15SE +/- 0.91, N = 15SE +/- 0.18, N = 15SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 1512.749.3811.128.728.985.6310.448.0010.177.467.9910.097.719.4022.0016.7922.5816.2117.8914.298.489.39

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.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: AES-128-GCMXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P500000M1000000M1500000M2000000M2500000MSE +/- 1622921637.03, N = 3SE +/- 1253503549.61, N = 3SE +/- 159598753.20, N = 3SE +/- 28627701.54, N = 3SE +/- 266669781.18, N = 3SE +/- 706172241.68, N = 3SE +/- 697208281.49, N = 3SE +/- 56355870.99, N = 3SE +/- 268465891.11, N = 3SE +/- 43545680.66, N = 3SE +/- 333650111.34, N = 3SE +/- 129279575.06, N = 3SE +/- 429428665.34, N = 3SE +/- 195953395.92, N = 3SE +/- 1841260363.24, N = 3SE +/- 49279694.68, N = 3SE +/- 1428966819.54, N = 3SE +/- 570058274.06, N = 3SE +/- 337702856.52, N = 3SE +/- 164763342.54, N = 3SE +/- 409615403.40, N = 3SE +/- 530918594.60, N = 317693131460138895675065401533026300727768804693697807229265300399526815427148466962705074771070266012775833827576412008481701266882866653638453203763109608219971356572076778323166119452831167221141870199035205996799600710926717850068790678972884361234889653908575397849313731. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: AES-256-GCMXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P400000M800000M1200000M1600000M2000000MSE +/- 1320809584.03, N = 3SE +/- 3966858436.56, N = 3SE +/- 1397764228.85, N = 3SE +/- 2848140790.67, N = 3SE +/- 116420153.53, N = 3SE +/- 81945775.93, N = 3SE +/- 112278705.55, N = 3SE +/- 63026881.96, N = 3SE +/- 108288170.03, N = 3SE +/- 119857064.56, N = 3SE +/- 301551944.80, N = 3SE +/- 317370827.95, N = 3SE +/- 252081360.30, N = 3SE +/- 34244051.67, N = 3SE +/- 1791964708.17, N = 3SE +/- 290015345.25, N = 3SE +/- 1859275059.09, N = 3SE +/- 671752057.50, N = 3SE +/- 2225250504.62, N = 3SE +/- 818689705.82, N = 3SE +/- 3442274884.66, N = 3SE +/- 148092414.72, N = 3149752993937774438768367013695343284707002393527177124854114473560620408831181148957353592014115453101189899332350745672808711253730411735689375783139794193887975063013746231983103026700997539007873171406176108785972059996315265054658137676965536674158939632074623097505301. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P300000M600000M900000M1200000M1500000MSE +/- 303075210.03, N = 3SE +/- 3335467.26, N = 3SE +/- 271318554.29, N = 3SE +/- 232288636.76, N = 3SE +/- 352722951.98, N = 3SE +/- 151152302.42, N = 3SE +/- 863978337.15, N = 3SE +/- 312824840.64, N = 3SE +/- 329510485.23, N = 3SE +/- 415487944.57, N = 3SE +/- 63366974.35, N = 3SE +/- 39095934.62, N = 3SE +/- 11835135.62, N = 3SE +/- 11028602.79, N = 3SE +/- 63368290.15, N = 3SE +/- 170968063.89, N = 3SE +/- 874940734.18, N = 3SE +/- 94935214.36, N = 3SE +/- 80233613.02, N = 3SE +/- 151272595.84, N = 3SE +/- 71110585.03, N = 3SE +/- 9570566.28, N = 36842843827603441509169376244029729473161264211734254511900632122948838305229124033472637406900674499084949272265973541533919112254801969206832703528268216531772490621831306928507573656396329387108661819195054490408477310015026348605033110211232820533270773070689383771. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: SHA512Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20000M40000M60000M80000M100000MSE +/- 14734422.48, N = 3SE +/- 37654528.63, N = 3SE +/- 481040738.21, N = 3SE +/- 203213936.42, N = 3SE +/- 20492408.59, N = 3SE +/- 6313493.56, N = 3SE +/- 14339684.98, N = 3SE +/- 1250834.62, N = 3SE +/- 3500318.33, N = 3SE +/- 500176.69, N = 3SE +/- 64065176.93, N = 3SE +/- 24045021.87, N = 3SE +/- 66704467.17, N = 3SE +/- 14430784.68, N = 3SE +/- 17716195.81, N = 3SE +/- 5822493.86, N = 3SE +/- 44023374.36, N = 3SE +/- 55712209.05, N = 3SE +/- 37307640.51, N = 3SE +/- 6285194.15, N = 3SE +/- 9990938.84, N = 3SE +/- 15082903.45, N = 34652076070723886054770428462544772221992772322079539900109809400933756452205318893980783323042229931619926127789194416857450887556407859939523040605137923104776633047527466341139145597120045865131913788222573833962757549722194470973243784608201. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20-Poly1305Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P200000M400000M600000M800000M1000000MSE +/- 82161991.11, N = 3SE +/- 120899798.30, N = 3SE +/- 132227134.30, N = 3SE +/- 766713911.50, N = 3SE +/- 31274617.76, N = 3SE +/- 255765711.98, N = 3SE +/- 36941390.18, N = 3SE +/- 37795783.50, N = 3SE +/- 4148679.28, N = 3SE +/- 6370607.14, N = 3SE +/- 38565703.74, N = 3SE +/- 18243420.30, N = 3SE +/- 37182497.08, N = 3SE +/- 2360978.39, N = 3SE +/- 19082587.00, N = 3SE +/- 151303743.70, N = 3SE +/- 558033954.71, N = 3SE +/- 4478378.70, N = 3SE +/- 6119305.11, N = 3SE +/- 29295317.86, N = 3SE +/- 65233629.47, N = 3SE +/- 24234601.89, N = 34440991442702236257380134422025329772250388507302935021224131461562129503727352152971877961744403203967204471613210582132665838578871340330390902400051351531205917872579060286089904570397488707676527195473839256520776989258222233509609983571933050936432111856617271. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

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.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-50Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.30, N = 15SE +/- 0.19, N = 3SE +/- 0.21, N = 3SE +/- 0.12, N = 3SE +/- 0.38, N = 15SE +/- 0.17, N = 3SE +/- 0.34, N = 15SE +/- 0.43, N = 15SE +/- 0.26, N = 3SE +/- 0.72, N = 3SE +/- 0.13, N = 9SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.24, N = 3SE +/- 0.22, N = 12SE +/- 0.04, N = 3SE +/- 0.29, N = 12SE +/- 0.19, N = 3SE +/- 0.23, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 340.8363.3037.5556.4437.4548.9937.8451.8641.6551.0515.9730.8714.8126.9113.7627.5316.8234.6916.2434.6533.1833.76

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.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.17, N = 12SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 15SE +/- 0.06, N = 15SE +/- 0.16, N = 15SE +/- 0.05, N = 3SE +/- 0.22, N = 12SE +/- 0.03, N = 3SE +/- 0.08, N = 15SE +/- 0.06, N = 3SE +/- 0.10, N = 15SE +/- 0.06, N = 3SE +/- 0.08, N = 15SE +/- 0.16, N = 3SE +/- 0.12, N = 15SE +/- 0.04, N = 3SE +/- 0.11, N = 15SE +/- 0.10, N = 7SE +/- 0.02, N = 3SE +/- 0.02, N = 38.888.068.827.106.434.297.416.826.816.306.748.476.417.909.6512.4110.6112.4310.5611.346.937.38

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.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-152Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.19, N = 5SE +/- 0.08, N = 3SE +/- 0.16, N = 4SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.12, N = 3SE +/- 0.13, N = 8SE +/- 0.16, N = 3SE +/- 0.18, N = 3SE +/- 0.26, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.17, N = 9SE +/- 0.12, N = 3SE +/- 0.15, N = 12SE +/- 0.04, N = 3SE +/- 0.20, N = 12SE +/- 0.01, N = 3SE +/- 0.27, N = 9SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.17, N = 317.5727.2115.7719.4116.6822.2714.8618.0717.9119.997.1014.536.4713.396.6413.058.3515.868.2715.5715.1115.25

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.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: GhostRider - Hash Count: 1MXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P7K14K21K28K35KSE +/- 64.69, N = 3SE +/- 44.68, N = 3SE +/- 4.03, N = 3SE +/- 23.16, N = 3SE +/- 49.14, N = 12SE +/- 32.20, N = 2SE +/- 7.44, N = 3SE +/- 14.78, N = 3SE +/- 14.02, N = 3SE +/- 12.93, N = 3SE +/- 43.14, N = 3SE +/- 43.00, N = 3SE +/- 53.95, N = 3SE +/- 19.74, N = 3SE +/- 46.94, N = 3SE +/- 61.27, N = 3SE +/- 95.10, N = 3SE +/- 10.36, N = 3SE +/- 290.50, N = 3SE +/- 117.32, N = 3SE +/- 41.94, N = 3SE +/- 30.53, N = 316324.78378.311263.25808.85230.92593.212115.06122.310975.85446.811147.25753.910431.65315.717873.19320.531157.815786.626702.714207.74564.34660.01. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P816243240SE +/- 0.30, N = 15SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.28, N = 15SE +/- 0.04, N = 3SE +/- 0.16, N = 15SE +/- 0.10, N = 3SE +/- 0.46, N = 15SE +/- 0.06, N = 3SE +/- 0.46, N = 12SE +/- 0.19, N = 3SE +/- 1.12, N = 15SE +/- 0.07, N = 3SE +/- 1.06, N = 15SE +/- 0.28, N = 5SE +/- 0.84, N = 15SE +/- 0.18, N = 15SE +/- 0.05, N = 3SE +/- 0.05, N = 319.9114.7418.7013.4614.318.8216.6412.5016.9111.8617.4515.8316.8815.9032.2623.7633.0824.6929.6121.3213.0013.81

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: particle_volume/scivis/real_timeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1326395265SE +/- 0.04682, N = 3SE +/- 0.04057, N = 3SE +/- 0.02499, N = 3SE +/- 0.00525, N = 3SE +/- 0.03711, N = 3SE +/- 0.00909, N = 3SE +/- 0.04134, N = 3SE +/- 0.03593, N = 3SE +/- 0.05872, N = 3SE +/- 0.00932, N = 3SE +/- 0.02131, N = 3SE +/- 0.01460, N = 3SE +/- 0.04097, N = 3SE +/- 0.00610, N = 3SE +/- 0.01106, N = 3SE +/- 0.02252, N = 3SE +/- 0.01792, N = 3SE +/- 0.01858, N = 3SE +/- 0.04960, N = 3SE +/- 0.01265, N = 3SE +/- 0.00583, N = 3SE +/- 0.01395, N = 353.9348030.6799038.2269021.8932017.622209.7002233.1275018.4851029.3496016.0869036.2464021.5147032.7979019.4251056.4485035.1464055.2576032.5232047.5267028.0689016.4505017.94480

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.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read While WritingXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P6M12M18M24M30MSE +/- 215352.04, N = 15SE +/- 28201.44, N = 3SE +/- 343737.14, N = 15SE +/- 61518.27, N = 10SE +/- 113154.86, N = 4SE +/- 23120.76, N = 3SE +/- 254193.37, N = 15SE +/- 26369.56, N = 3SE +/- 218479.19, N = 15SE +/- 83772.27, N = 13SE +/- 726178.20, N = 12SE +/- 136761.19, N = 4SE +/- 426084.70, N = 15SE +/- 45015.86, N = 3SE +/- 273109.47, N = 3SE +/- 109424.14, N = 3SE +/- 236709.61, N = 3SE +/- 93669.97, N = 15SE +/- 506458.47, N = 15SE +/- 36524.26, N = 3SE +/- 81998.56, N = 3SE +/- 28098.62, N = 3124255198926893123213007841738983668955398701122332973665009624417743966327059786118648152446614311340389229001521577618619851664141584192010828914608544893417693336451. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1122334455SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.35, N = 3SE +/- 0.01, N = 3SE +/- 0.33, N = 3SE +/- 0.31, N = 3SE +/- 0.40, N = 3SE +/- 0.11, N = 3SE +/- 0.35, N = 3SE +/- 0.06, N = 3SE +/- 0.36, N = 11SE +/- 0.08, N = 3SE +/- 0.23, N = 15SE +/- 0.45, N = 15SE +/- 0.40, N = 15SE +/- 0.61, N = 15SE +/- 0.61, N = 15SE +/- 0.50, N = 15SE +/- 0.61, N = 15SE +/- 0.63, N = 1528.8529.0229.7529.4129.7628.0236.0033.9936.2333.3343.8136.7847.2041.0827.9026.4522.7921.5825.3223.9226.8327.931. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1122334455SE +/- 0.11, N = 3SE +/- 0.10, N = 3SE +/- 0.13, N = 3SE +/- 0.05, N = 3SE +/- 0.40, N = 3SE +/- 0.01, N = 3SE +/- 0.26, N = 3SE +/- 0.27, N = 3SE +/- 0.31, N = 3SE +/- 0.09, N = 3SE +/- 0.18, N = 3SE +/- 0.05, N = 3SE +/- 0.16, N = 11SE +/- 0.05, N = 3SE +/- 0.27, N = 15SE +/- 0.63, N = 15SE +/- 0.76, N = 15SE +/- 1.36, N = 15SE +/- 0.92, N = 15SE +/- 0.91, N = 15SE +/- 0.82, N = 15SE +/- 0.87, N = 1534.6634.4633.6134.0033.6035.6927.7829.4327.6130.0022.8327.1821.2024.3435.8737.9644.0646.8839.8042.0737.5336.081. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Execution TimeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2004006008001000137.04302.06196.50420.65362.59798.35186.33410.29201.11440.83137.11449.4793.57178.25124.30325.42690.74691.641. (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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh TimeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200139.62148.40162.03160.73162.32182.24159.31150.67134.78153.12192.42167.41135.52118.54140.48136.40188.37163.701. (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

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P60120180240300SE +/- 0.27, N = 3SE +/- 0.12, N = 3SE +/- 0.15, N = 3SE +/- 0.07, N = 3SE +/- 0.37, N = 3SE +/- 1.03, N = 3SE +/- 0.79, N = 3SE +/- 0.36, N = 3SE +/- 0.17, N = 3SE +/- 0.19, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 1.52, N = 3SE +/- 0.20, N = 3SE +/- 0.45, N = 3SE +/- 0.74, N = 3SE +/- 0.09, N = 3SE +/- 0.27, N = 3SE +/- 0.08, N = 3SE +/- 0.31, N = 3SE +/- 0.10, N = 3SE +/- 0.20, N = 3106.82165.05122.31187.02194.38298.41128.07197.84134.42212.48125.27169.45138.57184.20101.56146.7689.28124.4694.81135.67215.93205.01

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.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2M4M6M8M10MSE +/- 35300.21, N = 3SE +/- 35182.67, N = 3SE +/- 50660.01, N = 15SE +/- 25335.31, N = 3SE +/- 101903.23, N = 15SE +/- 38410.91, N = 15SE +/- 24083.78, N = 3SE +/- 52575.97, N = 3SE +/- 26891.33, N = 3SE +/- 21428.06, N = 3SE +/- 16926.98, N = 15SE +/- 19998.04, N = 15SE +/- 16910.46, N = 15SE +/- 19070.02, N = 15SE +/- 14591.70, N = 3SE +/- 38372.96, N = 3SE +/- 27143.58, N = 3SE +/- 58081.25, N = 3SE +/- 83562.82, N = 3SE +/- 29027.89, N = 3SE +/- 41123.40, N = 3SE +/- 36366.89, N = 33780240.564752673.793245908.404481667.954814941.534826284.803392842.384315467.173373323.924384040.521821144.002339570.281783869.962299150.517645325.707736018.748683410.798634258.398498306.788220895.455820618.186198256.721. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

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.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.01358, N = 3SE +/- 0.02572, N = 3SE +/- 0.03225, N = 3SE +/- 0.01810, N = 3SE +/- 0.01280, N = 3SE +/- 0.01476, N = 3SE +/- 0.06816, N = 3SE +/- 0.03815, N = 5SE +/- 0.09530, N = 15SE +/- 0.02187, N = 3SE +/- 0.00701, N = 3SE +/- 0.03249, N = 3SE +/- 0.06840, N = 3SE +/- 0.08614, N = 3SE +/- 0.10131, N = 15SE +/- 0.07891, N = 15SE +/- 0.14799, N = 15SE +/- 0.01598, N = 3SE +/- 0.20624, N = 12SE +/- 0.08822, N = 15SE +/- 0.05392, N = 15SE +/- 0.04633, N = 155.839543.183925.625173.170024.493172.983066.177963.370345.744913.0289711.482006.9750912.434507.689778.518086.227917.096824.109147.555584.461354.742214.570291. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P70140210280350SE +/- 0.40, N = 3SE +/- 2.52, N = 3SE +/- 1.01, N = 3SE +/- 1.80, N = 3SE +/- 0.64, N = 3SE +/- 1.65, N = 3SE +/- 1.78, N = 3SE +/- 3.31, N = 5SE +/- 2.72, N = 15SE +/- 2.40, N = 3SE +/- 0.05, N = 3SE +/- 0.67, N = 3SE +/- 0.44, N = 3SE +/- 1.47, N = 3SE +/- 1.44, N = 15SE +/- 2.12, N = 15SE +/- 3.21, N = 15SE +/- 0.95, N = 3SE +/- 3.84, N = 12SE +/- 4.35, N = 15SE +/- 2.40, N = 15SE +/- 2.24, N = 15171.23314.06177.76315.41222.52335.15161.89296.80174.68330.1087.09143.3680.42130.06117.62160.90141.83243.34133.49225.31211.19219.061. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0b6Input: ATPase with 327,506 AtomsXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.34017, N = 12SE +/- 0.54361, N = 15SE +/- 0.17435, N = 15SE +/- 0.21567, N = 12SE +/- 0.07204, N = 4SE +/- 0.02582, N = 3SE +/- 0.44989, N = 15SE +/- 0.14537, N = 15SE +/- 0.26203, N = 15SE +/- 0.16058, N = 15SE +/- 0.20408, N = 12SE +/- 0.17734, N = 15SE +/- 0.19805, N = 15SE +/- 0.02593, N = 3SE +/- 0.02125, N = 3SE +/- 0.04917, N = 5SE +/- 0.08771, N = 3SE +/- 0.02148, N = 6SE +/- 0.04751, N = 3SE +/- 0.01368, N = 5SE +/- 0.00431, N = 4SE +/- 0.04338, N = 55.852674.648142.219482.970216.726633.604792.563195.925017.300655.971302.397703.763482.554003.7681118.974547.9975620.752599.3353417.811058.182755.449715.70672

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.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Read While WritingXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P8M16M24M32M40MSE +/- 195912.00, N = 15SE +/- 143570.67, N = 4SE +/- 239363.92, N = 3SE +/- 62510.72, N = 3SE +/- 48351.08, N = 3SE +/- 48302.92, N = 15SE +/- 197596.27, N = 3SE +/- 93731.24, N = 7SE +/- 171807.74, N = 3SE +/- 88257.34, N = 3SE +/- 223786.94, N = 15SE +/- 75416.94, N = 3SE +/- 138092.00, N = 3SE +/- 45758.35, N = 3SE +/- 803844.71, N = 15SE +/- 172052.60, N = 3SE +/- 651901.79, N = 15SE +/- 118870.12, N = 3SE +/- 746932.76, N = 15SE +/- 170909.18, N = 3SE +/- 86788.79, N = 3SE +/- 88013.69, N = 15175857371216841317297027106953601179531871471201535440099881761385925487954082765777617945745276307811754893436593204228294263330757419595777303245961802499311882103120539011. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-50Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.12, N = 3SE +/- 0.12, N = 3SE +/- 0.50, N = 3SE +/- 0.24, N = 3SE +/- 0.40, N = 15SE +/- 0.14, N = 3SE +/- 0.46, N = 4SE +/- 0.62, N = 3SE +/- 0.51, N = 3SE +/- 0.71, N = 3SE +/- 0.11, N = 3SE +/- 0.33, N = 5SE +/- 0.06, N = 3SE +/- 0.14, N = 3SE +/- 0.23, N = 9SE +/- 0.10, N = 3SE +/- 0.37, N = 12SE +/- 0.10, N = 3SE +/- 0.35, N = 12SE +/- 0.26, N = 3SE +/- 0.18, N = 3SE +/- 0.06, N = 340.7763.0637.0856.0137.7548.9938.3051.3542.3551.5516.1030.6914.8426.8913.6827.6817.2834.7117.0735.0133.2333.49

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.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.39, N = 13SE +/- 0.19, N = 3SE +/- 0.36, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.21, N = 3SE +/- 0.18, N = 3SE +/- 0.33, N = 3SE +/- 0.21, N = 3SE +/- 0.32, N = 15SE +/- 0.28, N = 3SE +/- 0.35, N = 3SE +/- 1.15, N = 3SE +/- 0.34, N = 3SE +/- 1.27, N = 15SE +/- 0.40, N = 15SE +/- 0.98, N = 15SE +/- 0.03, N = 3SE +/- 0.56, N = 15SE +/- 0.28, N = 4SE +/- 0.31, N = 15SE +/- 0.07, N = 326.0915.0427.5918.9321.9619.6029.4423.6624.6521.22167.9796.69179.14106.6955.9232.4742.8924.1142.5125.4832.2332.921. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1530456075SE +/- 0.50579, N = 13SE +/- 0.82440, N = 3SE +/- 0.47575, N = 3SE +/- 0.16471, N = 3SE +/- 0.18826, N = 3SE +/- 0.54692, N = 3SE +/- 0.20595, N = 3SE +/- 0.58786, N = 3SE +/- 0.34209, N = 3SE +/- 0.69010, N = 15SE +/- 0.00992, N = 3SE +/- 0.03724, N = 3SE +/- 0.03571, N = 3SE +/- 0.02972, N = 3SE +/- 0.40826, N = 15SE +/- 0.38216, N = 15SE +/- 0.56636, N = 15SE +/- 0.05911, N = 3SE +/- 0.31646, N = 15SE +/- 0.42374, N = 4SE +/- 0.29092, N = 15SE +/- 0.06539, N = 338.4240066.5162036.2493052.8167045.5414051.0172033.9621042.2747040.5671047.270305.9528310.342505.582419.3729418.0133030.8572023.4960041.4745023.5786039.2586031.0644030.374301. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.44, N = 3SE +/- 0.04, N = 3SE +/- 0.62, N = 3SE +/- 0.17, N = 3SE +/- 0.29, N = 3SE +/- 1.84, N = 12SE +/- 1.12, N = 15SE +/- 0.23, N = 3SE +/- 0.68, N = 3SE +/- 0.44, N = 3SE +/- 6.88, N = 15SE +/- 0.06, N = 3SE +/- 6.85, N = 15SE +/- 0.05, N = 3SE +/- 2.33, N = 12SE +/- 0.70, N = 15SE +/- 2.80, N = 15SE +/- 0.18, N = 3SE +/- 0.90, N = 6SE +/- 0.17, N = 3SE +/- 1.26, N = 3SE +/- 0.51, N = 344.6039.7560.3649.6454.2848.4755.9049.9760.1147.24141.57111.87155.90124.0693.6487.6690.0671.5693.8372.1286.6580.721. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.21982, N = 3SE +/- 0.02370, N = 3SE +/- 0.16743, N = 3SE +/- 0.07031, N = 3SE +/- 0.09869, N = 3SE +/- 0.58074, N = 12SE +/- 0.35497, N = 15SE +/- 0.09194, N = 3SE +/- 0.18711, N = 3SE +/- 0.19764, N = 3SE +/- 0.36771, N = 15SE +/- 0.00509, N = 3SE +/- 0.31004, N = 15SE +/- 0.00331, N = 3SE +/- 0.26635, N = 12SE +/- 0.09294, N = 15SE +/- 0.38561, N = 15SE +/- 0.03564, N = 3SE +/- 0.10002, N = 6SE +/- 0.03352, N = 3SE +/- 0.16639, N = 3SE +/- 0.07825, N = 322.4251025.1591016.5705020.1453018.4222020.8723017.9873020.0103016.6394021.170207.312858.938336.604718.0601310.7518811.4182011.2707013.9745010.6615013.8647011.5447012.388401. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.81431.62862.44293.25724.0715SE +/- 0.01541, N = 3SE +/- 0.00291, N = 3SE +/- 0.01405, N = 15SE +/- 0.01594, N = 3SE +/- 0.02289, N = 15SE +/- 0.00236, N = 3SE +/- 0.01853, N = 5SE +/- 0.01456, N = 3SE +/- 0.00498, N = 3SE +/- 0.02719, N = 15SE +/- 0.03971, N = 3SE +/- 0.00120, N = 3SE +/- 0.01119, N = 3SE +/- 0.00335, N = 3SE +/- 0.02347, N = 11SE +/- 0.01501, N = 3SE +/- 0.03659, N = 15SE +/- 0.01470, N = 3SE +/- 0.02209, N = 15SE +/- 0.02493, N = 15SE +/- 0.03117, N = 3SE +/- 0.02450, N = 41.324541.100421.435731.172921.428491.175611.718751.416621.652701.378313.415372.072883.619202.243233.164412.616532.246451.485962.250201.660462.238682.257161. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2004006008001000SE +/- 8.88, N = 3SE +/- 2.43, N = 3SE +/- 6.50, N = 15SE +/- 11.46, N = 3SE +/- 10.46, N = 15SE +/- 1.66, N = 3SE +/- 6.21, N = 5SE +/- 7.34, N = 3SE +/- 1.83, N = 3SE +/- 12.91, N = 15SE +/- 3.37, N = 3SE +/- 0.28, N = 3SE +/- 0.86, N = 3SE +/- 0.67, N = 3SE +/- 2.36, N = 11SE +/- 2.19, N = 3SE +/- 7.76, N = 15SE +/- 6.60, N = 3SE +/- 4.37, N = 15SE +/- 9.50, N = 15SE +/- 6.13, N = 3SE +/- 4.84, N = 4754.81908.19696.98852.33701.72849.56581.77705.61604.76728.47292.79482.21276.24445.62316.11382.10446.81672.92444.85604.05446.69443.061. (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.

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.7Test: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P246810SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.12, N = 6SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 33.054.083.825.853.104.411.541.541.771.472.675.582.565.394.637.475.248.485.248.355.115.16

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To CompileXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.29, N = 3SE +/- 0.91, N = 3SE +/- 0.56, N = 3SE +/- 0.76, N = 3SE +/- 0.69, N = 3SE +/- 0.22, N = 3SE +/- 1.69, N = 3SE +/- 0.32, N = 3SE +/- 0.56, N = 3SE +/- 0.18, N = 3SE +/- 1.40, N = 3SE +/- 0.73, N = 3SE +/- 0.64, N = 3SE +/- 0.16, N = 3SE +/- 0.55, N = 3SE +/- 0.20, N = 3SE +/- 0.62, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.16, N = 3SE +/- 0.16, N = 3SE +/- 0.32, N = 3103.38122.72118.57140.23151.54187.88122.70143.81123.88149.38148.99152.23160.46165.41126.11128.2896.16103.35102.58111.69159.62154.01

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.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-50Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.29, N = 3SE +/- 0.64, N = 3SE +/- 0.49, N = 3SE +/- 0.52, N = 3SE +/- 0.23, N = 3SE +/- 0.04, N = 3SE +/- 0.45, N = 3SE +/- 0.62, N = 3SE +/- 0.52, N = 3SE +/- 0.76, N = 3SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.28, N = 12SE +/- 0.08, N = 3SE +/- 0.32, N = 12SE +/- 0.09, N = 3SE +/- 0.37, N = 12SE +/- 0.21, N = 3SE +/- 0.13, N = 3SE +/- 0.18, N = 341.4563.6037.4454.8436.9748.8937.9151.5641.6852.9716.2230.8614.6726.8213.1127.7816.9734.9017.4134.9033.3033.68

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: INIVOL and Fluid Structure Interaction Drop ContainerXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.30, N = 3SE +/- 0.12, N = 3SE +/- 0.17, N = 3SE +/- 0.19, N = 3SE +/- 0.50, N = 3SE +/- 0.23, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.12, N = 3SE +/- 0.46, N = 3SE +/- 0.43, N = 3SE +/- 0.93, N = 3SE +/- 0.14, N = 3SE +/- 0.57, N = 3SE +/- 0.63, N = 3SE +/- 0.37, N = 3100.00128.74107.36140.43135.56204.21110.56146.03114.29160.5690.85106.5479.4294.7685.89100.24187.03176.87

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.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.17.1Input: Li2_STO_aeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.93, N = 3SE +/- 1.03, N = 3SE +/- 1.11, N = 3SE +/- 0.89, N = 3SE +/- 0.50, N = 3SE +/- 1.47, N = 7SE +/- 1.11, N = 5SE +/- 1.04, N = 3SE +/- 0.88, N = 3SE +/- 1.20, N = 3SE +/- 0.42, N = 3SE +/- 0.63, N = 3SE +/- 0.06, N = 3SE +/- 1.06, N = 3SE +/- 0.23, N = 3SE +/- 0.54, N = 3SE +/- 0.40, N = 3SE +/- 1.13, N = 397.91105.98101.63109.30131.37163.75106.27117.81102.12121.19111.15110.8091.9797.00102.98108.02151.27137.231. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -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.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P30060090012001500SE +/- 4.07, N = 3SE +/- 0.74, N = 3SE +/- 0.96, N = 3SE +/- 0.94, N = 3SE +/- 2.36, N = 15SE +/- 2.55, N = 3SE +/- 6.06, N = 3SE +/- 2.21, N = 3SE +/- 5.57, N = 13SE +/- 1.73, N = 3SE +/- 3.30, N = 3SE +/- 0.63, N = 3SE +/- 6.08, N = 15SE +/- 0.61, N = 3SE +/- 5.48, N = 3SE +/- 1.91, N = 3SE +/- 1.59, N = 3SE +/- 13.72, N = 15SE +/- 19.12, N = 12SE +/- 9.04, N = 12SE +/- 3.59, N = 15SE +/- 3.54, N = 3327.88302.27352.74329.04345.94363.35475.25382.11460.67380.90587.00515.08608.09562.40397.43319.351165.93801.961094.26800.23443.96388.741. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.74441.48882.23322.97763.722SE +/- 0.038257, N = 3SE +/- 0.008053, N = 3SE +/- 0.007683, N = 3SE +/- 0.008707, N = 3SE +/- 0.019617, N = 15SE +/- 0.019473, N = 3SE +/- 0.026516, N = 3SE +/- 0.015170, N = 3SE +/- 0.027020, N = 13SE +/- 0.011894, N = 3SE +/- 0.009611, N = 3SE +/- 0.002384, N = 3SE +/- 0.016386, N = 15SE +/- 0.001935, N = 3SE +/- 0.034519, N = 3SE +/- 0.018814, N = 3SE +/- 0.001170, N = 3SE +/- 0.020352, N = 15SE +/- 0.016597, N = 12SE +/- 0.013875, N = 12SE +/- 0.018603, N = 15SE +/- 0.023203, N = 33.0508503.3083602.8349803.0392202.8925302.7524202.1048002.6172102.1746302.6254401.7036501.9414401.6467401.7780902.5170903.1315600.8576851.2517900.9170431.2513502.2545302.5727801. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.00701, N = 3SE +/- 0.01491, N = 4SE +/- 0.01492, N = 3SE +/- 0.00936, N = 3SE +/- 0.00943, N = 3SE +/- 0.00184, N = 3SE +/- 0.00871, N = 3SE +/- 0.01053, N = 3SE +/- 0.00720, N = 3SE +/- 0.00884, N = 3SE +/- 0.11179, N = 3SE +/- 0.04327, N = 3SE +/- 0.07940, N = 3SE +/- 0.04869, N = 15SE +/- 0.03210, N = 15SE +/- 0.01454, N = 15SE +/- 0.04200, N = 3SE +/- 0.01586, N = 15SE +/- 0.02296, N = 15SE +/- 0.01428, N = 15SE +/- 0.01684, N = 6SE +/- 0.00301, N = 31.649191.202981.763531.273681.558211.301732.087801.472492.026591.4528012.126805.2405112.302105.664263.696022.210203.217811.693073.170641.705791.805731.824431. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2004006008001000SE +/- 2.57, N = 3SE +/- 10.06, N = 4SE +/- 4.81, N = 3SE +/- 5.72, N = 3SE +/- 3.90, N = 3SE +/- 1.09, N = 3SE +/- 1.98, N = 3SE +/- 4.87, N = 3SE +/- 1.77, N = 3SE +/- 4.16, N = 3SE +/- 0.77, N = 3SE +/- 1.57, N = 3SE +/- 0.53, N = 3SE +/- 1.50, N = 15SE +/- 2.41, N = 15SE +/- 3.02, N = 15SE +/- 4.10, N = 3SE +/- 5.53, N = 15SE +/- 2.35, N = 15SE +/- 4.99, N = 15SE +/- 5.34, N = 6SE +/- 0.91, N = 3605.59830.73566.38784.30641.00767.48478.47678.48492.89687.6782.44190.7881.26176.64270.63452.32310.59590.71315.36586.18553.47547.521. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P246810SE +/- 0.02371, N = 3SE +/- 0.00472, N = 3SE +/- 0.00316, N = 3SE +/- 0.00659, N = 3SE +/- 0.06148, N = 15SE +/- 0.03065, N = 15SE +/- 0.02608, N = 3SE +/- 0.01103, N = 3SE +/- 0.03283, N = 3SE +/- 0.08827, N = 15SE +/- 0.02282, N = 3SE +/- 0.00392, N = 3SE +/- 0.56391, N = 15SE +/- 0.00782, N = 3SE +/- 0.07124, N = 3SE +/- 0.06745, N = 3SE +/- 0.06302, N = 15SE +/- 0.01718, N = 3SE +/- 0.06624, N = 3SE +/- 0.03084, N = 3SE +/- 0.03354, N = 3SE +/- 0.06404, N = 132.802632.169663.465712.796943.431253.349665.391234.647575.022674.575814.748594.597036.143274.807257.840657.633373.807923.575435.984365.346567.860157.873251. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P100200300400500SE +/- 3.01, N = 3SE +/- 1.00, N = 3SE +/- 0.26, N = 3SE +/- 0.84, N = 3SE +/- 4.93, N = 15SE +/- 2.62, N = 15SE +/- 0.89, N = 3SE +/- 0.51, N = 3SE +/- 1.31, N = 3SE +/- 3.95, N = 15SE +/- 1.00, N = 3SE +/- 0.17, N = 3SE +/- 11.98, N = 15SE +/- 0.35, N = 3SE +/- 1.15, N = 3SE +/- 1.16, N = 3SE +/- 4.17, N = 15SE +/- 1.35, N = 3SE +/- 1.86, N = 3SE +/- 1.08, N = 3SE +/- 0.54, N = 3SE +/- 0.98, N = 13356.76460.78288.47357.36292.52298.67185.46215.13199.08219.53210.37217.45178.11207.94127.54131.00263.52279.64167.11187.01127.21127.091. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P60120180240300SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.14, N = 3SE +/- 0.92, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.10, N = 3SE +/- 0.18, N = 3SE +/- 0.34, N = 3SE +/- 0.15, N = 3SE +/- 0.07, N = 3SE +/- 0.25, N = 3SE +/- 1.90, N = 15SE +/- 6.97, N = 15SE +/- 0.09, N = 3SE +/- 1.22, N = 15SE +/- 0.44, N = 15SE +/- 2.90, N = 15SE +/- 0.41, N = 3SE +/- 0.19, N = 343.4139.6245.78100.39113.70236.8164.9159.6942.1641.93274.52266.91293.02283.8385.86230.2443.77110.9163.98157.45156.59161.481. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2004006008001000SE +/- 1.42, N = 3SE +/- 0.95, N = 3SE +/- 0.69, N = 3SE +/- 0.04, N = 3SE +/- 0.23, N = 3SE +/- 0.34, N = 3SE +/- 0.55, N = 3SE +/- 0.52, N = 3SE +/- 1.28, N = 3SE +/- 1.23, N = 3SE +/- 0.34, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 20.94, N = 15SE +/- 13.40, N = 15SE +/- 2.17, N = 3SE +/- 5.20, N = 15SE +/- 5.47, N = 15SE +/- 7.13, N = 15SE +/- 0.56, N = 3SE +/- 0.25, N = 3736.23403.20654.84298.56175.7284.30569.60301.30568.81285.95261.64134.77245.27126.64750.96283.181095.79433.06750.11306.27204.05197.911. (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.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.15083, N = 15SE +/- 0.01631, N = 3SE +/- 0.03218, N = 3SE +/- 0.03075, N = 3SE +/- 0.08423, N = 15SE +/- 0.04594, N = 3SE +/- 0.06105, N = 3SE +/- 0.06620, N = 3SE +/- 0.02902, N = 3SE +/- 0.06596, N = 15SE +/- 0.80628, N = 15SE +/- 0.02008, N = 3SE +/- 0.02533, N = 3SE +/- 0.02403, N = 3SE +/- 0.20120, N = 15SE +/- 0.01807, N = 3SE +/- 0.01742, N = 3SE +/- 0.02661, N = 3SE +/- 0.01221, N = 3SE +/- 0.08751, N = 5SE +/- 0.07964, N = 3SE +/- 0.03478, N = 34.295313.696495.551014.499765.164804.804187.532595.919617.041115.6770010.585716.746307.326107.1644011.5774011.139206.171906.166689.746147.8858411.2864011.204701. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P60120180240300SE +/- 8.14, N = 15SE +/- 1.19, N = 3SE +/- 1.05, N = 3SE +/- 1.50, N = 3SE +/- 3.16, N = 15SE +/- 1.97, N = 3SE +/- 1.08, N = 3SE +/- 1.88, N = 3SE +/- 0.58, N = 3SE +/- 2.01, N = 15SE +/- 8.91, N = 15SE +/- 0.44, N = 3SE +/- 0.47, N = 3SE +/- 0.47, N = 3SE +/- 1.43, N = 15SE +/- 0.15, N = 3SE +/- 0.46, N = 3SE +/- 0.70, N = 3SE +/- 0.13, N = 3SE +/- 1.37, N = 5SE +/- 0.63, N = 3SE +/- 0.28, N = 3236.65270.33180.03222.03194.15207.93132.71168.86141.95176.33103.89148.11136.39139.4586.6989.74161.95162.09102.57126.8288.5789.211. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 256 - Model: GoogLeNetXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P110220330440550SE +/- 2.68, N = 3SE +/- 2.30, N = 3SE +/- 4.87, N = 3SE +/- 0.24, N = 3SE +/- 0.69, N = 3SE +/- 0.16, N = 3SE +/- 4.80, N = 3SE +/- 1.91, N = 3SE +/- 2.29, N = 3SE +/- 1.67, N = 3SE +/- 2.49, N = 3SE +/- 0.07, N = 3SE +/- 3.73, N = 4SE +/- 0.31, N = 3SE +/- 1.11, N = 3SE +/- 10.93, N = 12SE +/- 0.51, N = 3SE +/- 2.76, N = 15SE +/- 1.02, N = 3SE +/- 3.87, N = 15SE +/- 0.07, N = 3SE +/- 0.04, N = 3501.39423.59528.93448.46402.78290.27449.91386.00447.50375.73335.38281.62301.46260.83397.44462.79391.40406.58401.84447.61287.42308.46

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: particle_volume/ao/real_timeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1326395265SE +/- 0.07203, N = 3SE +/- 0.01144, N = 3SE +/- 0.02269, N = 3SE +/- 0.00975, N = 3SE +/- 0.03319, N = 3SE +/- 0.05948, N = 3SE +/- 0.00297, N = 3SE +/- 0.01050, N = 3SE +/- 0.01945, N = 3SE +/- 0.03513, N = 3SE +/- 0.03595, N = 3SE +/- 0.01225, N = 3SE +/- 0.03817, N = 3SE +/- 0.02488, N = 3SE +/- 0.01800, N = 3SE +/- 0.03039, N = 3SE +/- 0.02899, N = 3SE +/- 0.01754, N = 3SE +/- 0.01406, N = 3SE +/- 0.01424, N = 3SE +/- 0.04390, N = 3SE +/- 0.02190, N = 354.1195030.8404038.2235021.9324017.826909.7776433.1125018.5533029.5018016.0502036.5043021.6222032.9663019.5159056.6936035.2461055.3382032.5188047.6143028.0956016.4504017.96680

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.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: yolov4 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.35, N = 3SE +/- 0.10, N = 3SE +/- 0.28, N = 3SE +/- 0.12, N = 3SE +/- 0.25, N = 3SE +/- 0.38, N = 3SE +/- 0.47, N = 14SE +/- 0.55, N = 15SE +/- 1.43, N = 15SE +/- 0.55, N = 15SE +/- 0.25, N = 3SE +/- 0.51, N = 3SE +/- 1.08, N = 3SE +/- 0.37, N = 3SE +/- 2.08, N = 3SE +/- 0.77, N = 3SE +/- 1.60, N = 3SE +/- 1.34, N = 3SE +/- 1.03, N = 3SE +/- 0.88, N = 8SE +/- 1.32, N = 5SE +/- 1.36, N = 458.9443.5464.2950.3560.9349.3969.2059.1869.4554.96122.7572.31133.3678.73148.21120.56137.99102.73155.96101.47114.31112.731. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: yolov4 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.09991, N = 3SE +/- 0.05084, N = 3SE +/- 0.06825, N = 3SE +/- 0.04802, N = 3SE +/- 0.06818, N = 3SE +/- 0.15605, N = 3SE +/- 0.09595, N = 14SE +/- 0.15263, N = 15SE +/- 0.29451, N = 15SE +/- 0.16821, N = 15SE +/- 0.01654, N = 3SE +/- 0.09768, N = 3SE +/- 0.06011, N = 3SE +/- 0.05969, N = 3SE +/- 0.09575, N = 3SE +/- 0.05314, N = 3SE +/- 0.08454, N = 3SE +/- 0.12519, N = 3SE +/- 0.04207, N = 3SE +/- 0.08554, N = 8SE +/- 0.09781, N = 5SE +/- 0.10588, N = 416.9672022.9689015.5555019.8593016.4123020.2495014.4583016.9166014.4831018.217908.1461513.829507.4994012.701506.749778.295337.248699.736856.412299.860188.752338.874461. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P48121620SE +/- 0.05342, N = 3SE +/- 0.00044, N = 3SE +/- 0.10855, N = 15SE +/- 0.00313, N = 3SE +/- 0.28946, N = 12SE +/- 0.04501, N = 3SE +/- 0.02558, N = 3SE +/- 0.00532, N = 3SE +/- 0.05347, N = 4SE +/- 0.00187, N = 3SE +/- 0.14238, N = 15SE +/- 0.07241, N = 3SE +/- 0.07602, N = 3SE +/- 0.08782, N = 3SE +/- 0.26893, N = 15SE +/- 0.02331, N = 3SE +/- 0.18371, N = 12SE +/- 0.03798, N = 3SE +/- 0.06617, N = 3SE +/- 0.04482, N = 3SE +/- 0.02847, N = 3SE +/- 0.02616, N = 33.980133.857744.725304.396495.298445.054275.002224.672694.858184.5891215.1274015.1057016.8062016.8631010.231069.421778.399698.1332610.434408.150929.711319.386811. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: StandardXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P60120180240300SE +/- 3.39, N = 3SE +/- 0.03, N = 3SE +/- 4.15, N = 15SE +/- 0.16, N = 3SE +/- 6.83, N = 12SE +/- 1.75, N = 3SE +/- 1.02, N = 3SE +/- 0.25, N = 3SE +/- 2.30, N = 4SE +/- 0.09, N = 3SE +/- 0.60, N = 15SE +/- 0.32, N = 3SE +/- 0.27, N = 3SE +/- 0.31, N = 3SE +/- 2.46, N = 15SE +/- 0.26, N = 3SE +/- 2.15, N = 12SE +/- 0.58, N = 3SE +/- 0.61, N = 3SE +/- 0.67, N = 3SE +/- 0.30, N = 3SE +/- 0.30, N = 3251.31259.19212.92227.42192.81197.85199.90213.99205.89217.8866.1866.2059.5059.3098.64106.13119.56122.9595.84122.68102.97106.531. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20K40K60K80K100KSE +/- 16.05, N = 3SE +/- 90.37, N = 3SE +/- 64.24, N = 3SE +/- 102.40, N = 3SE +/- 152.29, N = 3SE +/- 175.68, N = 3SE +/- 13.86, N = 3SE +/- 64.91, N = 3SE +/- 37.55, N = 3SE +/- 134.33, N = 3SE +/- 605.82, N = 15SE +/- 424.84, N = 3SE +/- 739.02, N = 15SE +/- 541.85, N = 3SE +/- 47.34, N = 3SE +/- 312.83, N = 12SE +/- 14.62, N = 3SE +/- 193.23, N = 3SE +/- 109.43, N = 3SE +/- 42.05, N = 3SE +/- 346.19, N = 3SE +/- 172.59, N = 3242495079833411580265524710346136179625403892668973560828100362449885791696037383178553864520750432907809372051

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P30K60K90K120K150KSE +/- 47.16, N = 3SE +/- 162.52, N = 3SE +/- 125.33, N = 3SE +/- 58.68, N = 3SE +/- 32.38, N = 3SE +/- 623.29, N = 3SE +/- 141.04, N = 3SE +/- 65.97, N = 3SE +/- 155.80, N = 3SE +/- 41.33, N = 3SE +/- 176.25, N = 3SE +/- 189.03, N = 3SE +/- 900.17, N = 15SE +/- 365.52, N = 3SE +/- 28.11, N = 3SE +/- 50.97, N = 3SE +/- 27.74, N = 3SE +/- 89.57, N = 3SE +/- 756.08, N = 15SE +/- 231.95, N = 3SE +/- 13.58, N = 3SE +/- 29.36, N = 33339159522388136800664188121734419717336645211807426648792175715831014442007142945208934434927943500199164384261

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.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Decompression SpeedXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P30060090012001500SE +/- 1.88, N = 3SE +/- 3.01, N = 4SE +/- 4.96, N = 3SE +/- 4.45, N = 3SE +/- 1.74, N = 4SE +/- 1.75, N = 3SE +/- 0.65, N = 15SE +/- 1.08, N = 15SE +/- 0.66, N = 15SE +/- 2.12, N = 3SE +/- 2.25, N = 3SE +/- 5.13, N = 3SE +/- 5.06, N = 3SE +/- 2.22, N = 3SE +/- 1.83, N = 3SE +/- 0.83, N = 3SE +/- 1.73, N = 3SE +/- 7.31, N = 3SE +/- 2.20, N = 3SE +/- 0.46, N = 3SE +/- 1.58, N = 3SE +/- 21.23, N = 31165.81221.8963.61032.0900.51032.41039.21127.51040.21126.4774.4816.7721.4743.51139.01139.91392.51385.31382.11388.91144.81118.81. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Compression SpeedXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P48121620SE +/- 0.03, N = 3SE +/- 0.21, N = 4SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.17, N = 4SE +/- 0.03, N = 3SE +/- 0.13, N = 15SE +/- 0.11, N = 15SE +/- 0.09, N = 15SE +/- 0.15, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.22, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 317.818.015.214.614.313.713.314.513.814.411.611.510.610.613.713.816.817.217.017.213.813.91. (CC) gcc options: -O3 -pthread -lz -llzma

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.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Random ReadXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P300M600M900M1200M1500MSE +/- 5399265.88, N = 15SE +/- 242146.75, N = 3SE +/- 5393159.45, N = 3SE +/- 168680.90, N = 3SE +/- 706582.63, N = 3SE +/- 232538.80, N = 3SE +/- 2955608.21, N = 3SE +/- 2763560.64, N = 13SE +/- 8491098.40, N = 15SE +/- 3698537.65, N = 15SE +/- 2674444.99, N = 3SE +/- 1322733.50, N = 3SE +/- 1272758.30, N = 3SE +/- 998087.60, N = 3SE +/- 2399164.12, N = 3SE +/- 1606748.01, N = 3SE +/- 5954964.38, N = 3SE +/- 205170.43, N = 3SE +/- 955915.20, N = 3SE +/- 66268.38, N = 3SE +/- 70537.88, N = 3SE +/- 109015.29, N = 358614787131517838155128695827617961725913646613035914549790749024407067441994357021145704813241633406731592171158496546596955494116994738558995107010595465335376330348576871794285570542429401102740917781. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.11Video Input: Bosphorus 4K - Video Preset: FastXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780EXeon 6766EEPYC 9754EPYC 9684XEPYC 9654EPYC 8534PNEPYC 8534P246810SE +/- 0.038, N = 3SE +/- 0.023, N = 3SE +/- 0.047, N = 3SE +/- 0.022, N = 3SE +/- 0.028, N = 3SE +/- 0.010, N = 3SE +/- 0.043, N = 15SE +/- 0.033, N = 3SE +/- 0.081, N = 3SE +/- 0.010, N = 3SE +/- 0.007, N = 3SE +/- 0.008, N = 3SE +/- 0.045, N = 3SE +/- 0.057, N = 3SE +/- 0.048, N = 3SE +/- 0.041, N = 3SE +/- 0.038, N = 36.9607.3926.4806.7195.8775.5836.0776.5016.0236.3244.4404.0526.4657.5477.4316.0006.1181. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

Video Input: Bosphorus 4K - Video Preset: Fast

EPYC 9684X 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1

EPYC 9654 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

Xeon 6766E 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1

Xeon 6780E 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-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.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Decompression SpeedXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P30060090012001500SE +/- 1.60, N = 3SE +/- 0.84, N = 3SE +/- 3.69, N = 3SE +/- 0.43, N = 3SE +/- 0.30, N = 3SE +/- 4.15, N = 3SE +/- 1.78, N = 15SE +/- 1.45, N = 3SE +/- 1.00, N = 3SE +/- 0.56, N = 3SE +/- 1.54, N = 3SE +/- 1.05, N = 3SE +/- 4.61, N = 3SE +/- 3.99, N = 3SE +/- 0.19, N = 3SE +/- 1.95, N = 3SE +/- 2.93, N = 3SE +/- 1.99, N = 3SE +/- 1.87, N = 3SE +/- 1.17, N = 3SE +/- 0.92, N = 3SE +/- 2.09, N = 31189.71242.6992.11069.1912.91018.31046.01133.31041.81131.1753.2797.7701.2739.11104.21102.21321.41324.91324.61328.51096.51101.41. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Compression SpeedXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 15SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 39.679.808.167.947.837.527.467.727.627.766.106.195.605.656.646.688.428.458.308.386.746.761. (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.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.3060.6120.9181.2241.53SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 15SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 9SE +/- 0.00, N = 3SE +/- 0.00, N = 7SE +/- 0.00, N = 3SE +/- 0.00, N = 15SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.460.490.290.430.450.560.300.370.280.331.091.251.201.360.410.730.350.450.350.480.600.531. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P40K80K120K160K200KSE +/- 1158.60, N = 3SE +/- 988.12, N = 3SE +/- 2494.14, N = 3SE +/- 75.30, N = 3SE +/- 696.52, N = 3SE +/- 95.54, N = 3SE +/- 1422.88, N = 15SE +/- 1761.33, N = 3SE +/- 18.38, N = 3SE +/- 383.22, N = 3SE +/- 1067.91, N = 3SE +/- 100.17, N = 3SE +/- 1048.23, N = 3SE +/- 68.06, N = 3SE +/- 1374.23, N = 9SE +/- 89.35, N = 3SE +/- 1640.61, N = 7SE +/- 162.47, N = 3SE +/- 1922.38, N = 15SE +/- 205.87, N = 3SE +/- 55.52, N = 3SE +/- 86.99, N = 3128566.8181279.79189893.31115106.30151627.4261956.93176900.32140144.01174862.56135075.0776895.2991273.0275761.5084584.13166151.42132973.43178488.03138605.52172688.50133779.6278334.9287511.111. (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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P7001400210028003500SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 4.62, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 1.76, N = 3SE +/- 0.33, N = 3SE +/- 24.01, N = 12SE +/- 11.06, N = 3SE +/- 21.79, N = 12SE +/- 7.84, N = 3SE +/- 0.88, N = 3SE +/- 1.20, N = 3SE +/- 1.20, N = 3SE +/- 1.15, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 1.20, N = 3SE +/- 0.58, N = 3759144189416571553305296117921049198813422090149723185319545611037661118022402043

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 1 - Model: ResNet-50Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P48121620SE +/- 0.03, N = 15SE +/- 0.05, N = 9SE +/- 0.06, N = 15SE +/- 0.05, N = 4SE +/- 0.13, N = 15SE +/- 0.08, N = 5SE +/- 0.04, N = 15SE +/- 0.10, N = 4SE +/- 0.07, N = 3SE +/- 0.09, N = 4SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 15SE +/- 0.04, N = 3SE +/- 0.03, N = 15SE +/- 0.05, N = 15SE +/- 0.01, N = 3SE +/- 0.09, N = 15SE +/- 0.06, N = 15SE +/- 0.01, N = 4SE +/- 0.08, N = 15SE +/- 0.08, N = 153.856.434.3910.539.1317.824.379.995.1111.291.995.021.764.672.476.193.928.963.669.0110.7010.79

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: ResNet-50Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.70, N = 3SE +/- 0.50, N = 3SE +/- 1.12, N = 3SE +/- 0.04, N = 3SE +/- 1.42, N = 3SE +/- 0.36, N = 3SE +/- 1.28, N = 3SE +/- 0.93, N = 3SE +/- 0.39, N = 3SE +/- 0.13, N = 3SE +/- 0.76, N = 3SE +/- 0.14, N = 3SE +/- 0.56, N = 3SE +/- 0.04, N = 3SE +/- 0.88, N = 3SE +/- 0.60, N = 3SE +/- 1.04, N = 4SE +/- 0.08, N = 3SE +/- 0.91, N = 5SE +/- 0.64, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 394.6691.51115.47120.93108.3290.72107.30103.56111.71104.2368.5276.3761.4871.4167.08106.1486.08104.5282.94114.6380.8687.31

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P60120180240300SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.16, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.39, N = 3SE +/- 0.05, N = 3SE +/- 0.34, N = 3SE +/- 0.07, N = 3SE +/- 0.12, N = 3SE +/- 3.14, N = 3SE +/- 0.06, N = 3SE +/- 2.84, N = 15SE +/- 0.48, N = 3SE +/- 0.59, N = 3SE +/- 1.05, N = 5SE +/- 0.01, N = 3SE +/- 0.60, N = 11SE +/- 0.44, N = 3SE +/- 0.77, N = 7SE +/- 0.05, N = 3SE +/- 0.13, N = 328.1726.8537.5862.4670.95127.8941.5738.5029.9828.73271.14240.38294.69259.6656.73102.8443.1783.1148.4391.16109.00100.671. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2004006008001000SE +/- 0.85, N = 3SE +/- 0.54, N = 3SE +/- 3.48, N = 3SE +/- 0.17, N = 3SE +/- 0.17, N = 3SE +/- 0.45, N = 3SE +/- 1.13, N = 3SE +/- 4.09, N = 3SE +/- 1.99, N = 3SE +/- 1.21, N = 3SE +/- 3.00, N = 3SE +/- 0.03, N = 3SE +/- 2.30, N = 15SE +/- 0.25, N = 3SE +/- 11.78, N = 3SE +/- 6.34, N = 5SE +/- 0.37, N = 3SE +/- 4.29, N = 11SE +/- 8.96, N = 3SE +/- 4.59, N = 7SE +/- 0.13, N = 3SE +/- 0.39, N = 31128.78592.20793.64477.77281.36156.13889.18467.08795.63415.96264.66149.63243.76138.531127.05621.421110.49576.98990.16525.88293.07317.381. (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.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P816243240SE +/- 0.06875, N = 3SE +/- 0.02630, N = 3SE +/- 0.02968, N = 3SE +/- 0.00544, N = 3SE +/- 0.03509, N = 3SE +/- 0.01736, N = 3SE +/- 0.03080, N = 3SE +/- 0.03023, N = 3SE +/- 0.01680, N = 3SE +/- 0.01698, N = 3SE +/- 0.32730, N = 6SE +/- 0.07856, N = 3SE +/- 0.34399, N = 3SE +/- 0.08701, N = 3SE +/- 0.08457, N = 3SE +/- 0.02258, N = 3SE +/- 0.05309, N = 3SE +/- 0.04669, N = 3SE +/- 0.05293, N = 3SE +/- 0.06505, N = 4SE +/- 0.03633, N = 15SE +/- 0.03641, N = 156.496933.237806.390313.192985.180703.173606.517443.320306.005253.1336635.2266015.5856036.4009017.9000011.314806.523679.952455.594919.962455.253214.926534.825061. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P70140210280350SE +/- 1.63, N = 3SE +/- 2.50, N = 3SE +/- 0.73, N = 3SE +/- 0.53, N = 3SE +/- 1.30, N = 3SE +/- 1.73, N = 3SE +/- 0.73, N = 3SE +/- 2.72, N = 3SE +/- 0.47, N = 3SE +/- 1.74, N = 3SE +/- 0.27, N = 6SE +/- 0.32, N = 3SE +/- 0.26, N = 3SE +/- 0.27, N = 3SE +/- 0.66, N = 3SE +/- 0.53, N = 3SE +/- 0.54, N = 3SE +/- 1.50, N = 3SE +/- 0.54, N = 3SE +/- 2.31, N = 4SE +/- 1.50, N = 15SE +/- 1.54, N = 15153.92308.80156.46313.10192.99315.03153.41301.14166.49319.0428.4064.1627.4755.8688.37153.25100.46178.70100.36190.38203.06207.341. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P70140210280350SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.86, N = 3SE +/- 0.65, N = 3SE +/- 4.48, N = 3SE +/- 0.66, N = 3SE +/- 0.10, N = 3SE +/- 0.45, N = 12SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.08, N = 3SE +/- 0.16, N = 14SE +/- 0.02, N = 3SE +/- 0.00, N = 313.159.4015.9613.9526.3325.9118.7212.7217.1511.37293.3675.22340.2682.1228.1022.6722.2013.5223.1216.3218.4917.301. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2K4K6K8K10KSE +/- 24.76, N = 3SE +/- 10.14, N = 3SE +/- 13.31, N = 3SE +/- 1.87, N = 3SE +/- 6.96, N = 3SE +/- 1.70, N = 3SE +/- 28.98, N = 3SE +/- 7.60, N = 3SE +/- 37.79, N = 3SE +/- 15.13, N = 3SE +/- 2.88, N = 3SE +/- 16.46, N = 3SE +/- 11.12, N = 3SE +/- 13.67, N = 3SE +/- 27.90, N = 3SE +/- 139.41, N = 12SE +/- 59.51, N = 3SE +/- 1.84, N = 3SE +/- 77.59, N = 3SE +/- 67.22, N = 14SE +/- 3.67, N = 3SE +/- 0.64, N = 39360.896757.267356.634209.102963.821525.575933.654386.485565.234208.49978.461906.99843.921747.506513.365310.078231.726963.317810.655697.513405.943636.331. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 0Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.32, N = 3SE +/- 0.18, N = 3SE +/- 0.52, N = 3SE +/- 0.32, N = 3SE +/- 0.22, N = 3SE +/- 0.29, N = 3SE +/- 0.15, N = 3SE +/- 0.18, N = 3SE +/- 0.12, N = 3SE +/- 0.22, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.13, N = 3SE +/- 0.01, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.01, N = 3SE +/- 0.18, N = 3SE +/- 0.19, N = 360.5364.4867.3869.9388.07100.4668.4473.0669.6576.92129.91126.97143.80140.8674.6973.2763.4061.6362.8961.9978.2475.861. (CXX) g++ options: -O3 -fPIC -lm

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.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.01381, N = 3SE +/- 0.01801, N = 3SE +/- 0.04801, N = 3SE +/- 0.00807, N = 3SE +/- 0.01782, N = 3SE +/- 0.02857, N = 3SE +/- 0.02633, N = 15SE +/- 0.01443, N = 3SE +/- 0.04072, N = 3SE +/- 0.03338, N = 15SE +/- 0.14331, N = 3SE +/- 0.01135, N = 3SE +/- 0.12596, N = 3SE +/- 0.01143, N = 3SE +/- 0.01019, N = 3SE +/- 0.02990, N = 3SE +/- 0.00852, N = 3SE +/- 0.01812, N = 3SE +/- 0.03309, N = 3SE +/- 0.00389, N = 3SE +/- 0.02993, N = 3SE +/- 0.03399, N = 32.817392.159804.058813.065004.006753.542423.958192.887344.052702.9709410.588304.0627210.806604.203553.987353.129043.617922.239223.680012.667113.924553.910061. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P100200300400500SE +/- 1.73, N = 3SE +/- 3.84, N = 3SE +/- 2.94, N = 3SE +/- 0.86, N = 3SE +/- 1.11, N = 3SE +/- 2.25, N = 3SE +/- 1.67, N = 15SE +/- 1.74, N = 3SE +/- 2.44, N = 3SE +/- 3.48, N = 15SE +/- 1.28, N = 3SE +/- 0.69, N = 3SE +/- 1.08, N = 3SE +/- 0.65, N = 3SE +/- 0.64, N = 3SE +/- 3.02, N = 3SE +/- 0.65, N = 3SE +/- 3.58, N = 3SE +/- 2.43, N = 3SE +/- 0.55, N = 3SE +/- 1.96, N = 3SE +/- 2.24, N = 3354.75462.86246.34326.14249.48282.25252.67346.21246.68337.0094.42246.0492.51237.80250.63319.45276.21446.30271.62374.75254.73255.671. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P30K60K90K120K150KSE +/- 1080.03, N = 3SE +/- 219.20, N = 3SE +/- 412.68, N = 4SE +/- 76.67, N = 3SE +/- 46.25, N = 3SE +/- 47.29, N = 3SE +/- 802.88, N = 3SE +/- 0.00, N = 3SE +/- 194.30, N = 3SE +/- 99.83, N = 3SE +/- 1180.69, N = 14SE +/- 685.49, N = 12SE +/- 19579.11, N = 3SE +/- 581.17, N = 3SE +/- 2386.93, N = 4SE +/- 1853.72, N = 3SE +/- 0.00, N = 3SE +/- 86.77, N = 372004.0037265.9041773.0021314.2014497.309623.7745195.1023250.2042939.2021766.5054066.0028854.30121976.9074389.0086757.7046233.7016399.7016573.201. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P30K60K90K120K150KSE +/- 480.95, N = 3SE +/- 153.57, N = 3SE +/- 378.22, N = 4SE +/- 93.91, N = 3SE +/- 62.13, N = 3SE +/- 64.71, N = 3SE +/- 0.00, N = 3SE +/- 76.47, N = 3SE +/- 143.60, N = 3SE +/- 0.00, N = 3SE +/- 1077.92, N = 14SE +/- 598.81, N = 12SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 882.91, N = 4SE +/- 189.10, N = 3SE +/- 69.57, N = 3SE +/- 44.43, N = 359139.2031177.2034614.1017891.6012800.408429.1340368.6020561.4036900.5018522.0052639.5028476.60123480.0064260.2081025.8042172.4015903.4016704.401. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digitXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.26, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.42, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.53, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.32, N = 3SE +/- 0.02, N = 3SE +/- 0.42, N = 3SE +/- 0.14, N = 3SE +/- 0.38, N = 5SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.27, N = 348.0989.0750.8393.3582.96165.7958.25107.2465.03123.8353.02103.8658.96115.4626.7848.8530.6956.9133.6862.17120.56105.681. (CC) gcc options: -O2 -pthread

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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P12K24K36K48K60KSE +/- 7.13, N = 3SE +/- 37.49, N = 3SE +/- 34.10, N = 3SE +/- 62.46, N = 3SE +/- 53.03, N = 3SE +/- 148.46, N = 3SE +/- 51.39, N = 3SE +/- 83.47, N = 3SE +/- 72.42, N = 3SE +/- 28.06, N = 3SE +/- 429.71, N = 4SE +/- 371.33, N = 3SE +/- 409.89, N = 15SE +/- 199.93, N = 3SE +/- 12.34, N = 3SE +/- 33.33, N = 3SE +/- 17.16, N = 3SE +/- 39.30, N = 3SE +/- 32.74, N = 3SE +/- 44.33, N = 3SE +/- 31.02, N = 3SE +/- 250.73, N = 3121242312514138313803044254456153573379016645370553508146875382285208384131528689161660310428188964251338811

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P14K28K42K56K70KSE +/- 8.00, N = 3SE +/- 26.00, N = 3SE +/- 29.96, N = 3SE +/- 61.20, N = 3SE +/- 28.75, N = 3SE +/- 62.75, N = 3SE +/- 10.97, N = 3SE +/- 74.33, N = 3SE +/- 25.12, N = 3SE +/- 36.47, N = 3SE +/- 301.00, N = 12SE +/- 26.58, N = 3SE +/- 51.90, N = 3SE +/- 500.31, N = 3SE +/- 16.90, N = 3SE +/- 21.05, N = 3SE +/- 32.79, N = 3SE +/- 24.66, N = 3SE +/- 27.79, N = 3SE +/- 5.55, N = 3SE +/- 197.99, N = 3SE +/- 111.72, N = 314358319931683736649348666303518111392381970142837392835250143906576571001518129105141953512427223054884445377

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeaf 1.3Input: clover_bm64_shortXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.45, N = 3SE +/- 0.03, N = 3SE +/- 0.24, N = 3SE +/- 0.02, N = 3SE +/- 0.31, N = 3SE +/- 0.03, N = 3SE +/- 0.40, N = 4SE +/- 0.17, N = 3SE +/- 0.31, N = 15SE +/- 0.15, N = 3SE +/- 0.33, N = 3SE +/- 0.04, N = 3SE +/- 0.33, N = 3SE +/- 0.03, N = 3SE +/- 0.30, N = 3SE +/- 2.07, N = 12SE +/- 0.23, N = 3SE +/- 1.81, N = 15SE +/- 1.11, N = 12SE +/- 2.15, N = 15SE +/- 0.04, N = 3SE +/- 0.04, N = 331.9039.3826.5839.4835.2960.8932.4442.2232.2541.4328.9238.0128.9938.0127.7448.3622.8342.1730.0546.5057.5457.381. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

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.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P8001600240032004000SE +/- 0.00, N = 3SE +/- 1.45, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 2.73, N = 3SE +/- 3.79, N = 3SE +/- 0.33, N = 3SE +/- 1.20, N = 3SE +/- 2.65, N = 3SE +/- 1.20, N = 3SE +/- 16.65, N = 5SE +/- 3.51, N = 3SE +/- 4.67, N = 3SE +/- 6.23, N = 3SE +/- 0.67, N = 3SE +/- 0.88, N = 3SE +/- 1.45, N = 3SE +/- 1.53, N = 3SE +/- 0.58, N = 3SE +/- 1.00, N = 3SE +/- 3.06, N = 3SE +/- 1.00, N = 3898170910631977183336161143212312372357163124721832273163011306601225780139126492421

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1530456075SE +/- 0.28, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.14, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.21, N = 3SE +/- 0.05, N = 3SE +/- 0.15, N = 3SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 3SE +/- 0.02, N = 3SE +/- 0.26, N = 15SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 357.1355.0765.0255.9916.8431.0562.1162.5356.7155.7437.1735.3041.5038.8715.8438.2111.9521.6314.3027.4634.7631.991. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P9001800270036004500SE +/- 11.53, N = 3SE +/- 0.47, N = 3SE +/- 0.66, N = 3SE +/- 2.65, N = 3SE +/- 3.75, N = 3SE +/- 0.62, N = 3SE +/- 1.74, N = 3SE +/- 2.94, N = 3SE +/- 1.56, N = 3SE +/- 2.29, N = 3SE +/- 2.23, N = 3SE +/- 2.82, N = 3SE +/- 2.08, N = 3SE +/- 2.69, N = 3SE +/- 5.40, N = 3SE +/- 11.41, N = 15SE +/- 5.33, N = 3SE +/- 1.09, N = 3SE +/- 3.90, N = 3SE +/- 6.85, N = 3SE +/- 0.81, N = 3SE +/- 0.65, N = 32237.691159.661843.611068.431186.24642.441801.89894.871691.79860.641933.081018.181731.50924.904032.661671.234013.962212.413353.731743.49918.53998.091. (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.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P60120180240300SE +/- 0.28, N = 3SE +/- 0.07, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.37, N = 3SE +/- 0.06, N = 3SE +/- 0.13, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.17, N = 3SE +/- 1.74, N = 15SE +/- 0.89, N = 3SE +/- 3.87, N = 3SE +/- 0.84, N = 3SE +/- 0.45, N = 3SE +/- 0.13, N = 3SE +/- 0.22, N = 3SE +/- 0.05, N = 3SE +/- 0.25, N = 3SE +/- 0.04, N = 3SE +/- 0.35, N = 3SE +/- 0.17, N = 331.7916.5534.3121.3425.9822.5934.2523.6630.4724.55266.10146.78285.99160.1383.7740.6366.4433.5066.6334.1638.1036.611. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.27122, N = 3SE +/- 0.27014, N = 3SE +/- 0.09239, N = 3SE +/- 0.09731, N = 3SE +/- 0.55506, N = 3SE +/- 0.10899, N = 3SE +/- 0.11212, N = 3SE +/- 0.20934, N = 3SE +/- 0.11901, N = 3SE +/- 0.27882, N = 3SE +/- 0.02474, N = 15SE +/- 0.04131, N = 3SE +/- 0.04782, N = 3SE +/- 0.03255, N = 3SE +/- 0.06474, N = 3SE +/- 0.07716, N = 3SE +/- 0.05013, N = 3SE +/- 0.04145, N = 3SE +/- 0.05569, N = 3SE +/- 0.03174, N = 3SE +/- 0.23936, N = 3SE +/- 0.12474, N = 331.4589060.4094029.1487046.8491038.5025044.2748029.1955042.2711032.8141040.728903.760036.813303.497856.2452111.9382024.6095015.0516029.8493015.0082029.2690026.2475027.312801. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.45, N = 15SE +/- 0.46, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.33, N = 3SE +/- 0.10, N = 3SE +/- 0.13, N = 3SE +/- 0.34, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.09, N = 322.9621.8725.1236.2060.9062.6425.7224.9424.0023.12157.28153.76180.29172.7239.8638.8533.1832.0036.9034.9245.7141.381. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P14002800420056007000SE +/- 3.43, N = 3SE +/- 0.73, N = 3SE +/- 2.46, N = 3SE +/- 1.46, N = 3SE +/- 9.58, N = 15SE +/- 4.60, N = 3SE +/- 2.81, N = 3SE +/- 4.55, N = 3SE +/- 1.09, N = 3SE +/- 3.41, N = 3SE +/- 3.81, N = 3SE +/- 0.61, N = 3SE +/- 1.20, N = 3SE +/- 1.73, N = 3SE +/- 14.68, N = 3SE +/- 7.27, N = 3SE +/- 13.59, N = 3SE +/- 4.06, N = 3SE +/- 9.83, N = 3SE +/- 4.84, N = 3SE +/- 0.47, N = 3SE +/- 3.51, N = 35570.632924.664744.051643.431313.39637.764351.592243.843998.022074.301826.82934.971593.88832.296400.943280.255766.622986.335178.292739.011396.781542.611. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.12, N = 15SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 314.6713.7321.7317.844.549.2620.2419.4417.1916.3314.3014.0116.2815.115.0010.974.087.854.468.5811.4810.781. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3K6K9K12K15KSE +/- 6.25, N = 3SE +/- 0.73, N = 3SE +/- 11.51, N = 3SE +/- 0.81, N = 3SE +/- 2.84, N = 3SE +/- 1.06, N = 3SE +/- 8.70, N = 3SE +/- 5.68, N = 3SE +/- 4.11, N = 3SE +/- 8.09, N = 3SE +/- 5.08, N = 3SE +/- 2.61, N = 3SE +/- 8.74, N = 3SE +/- 1.12, N = 3SE +/- 1.24, N = 3SE +/- 66.33, N = 15SE +/- 4.04, N = 3SE +/- 3.54, N = 3SE +/- 4.21, N = 3SE +/- 8.39, N = 3SE +/- 2.89, N = 3SE +/- 2.22, N = 38719.374656.925483.813325.904390.812120.335525.562877.325576.862935.474992.782553.554376.092366.8312750.595778.5111735.426017.1910708.915528.152763.482943.081. (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.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1326395265SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.30, N = 3SE +/- 0.32, N = 3SE +/- 0.08, N = 3SE +/- 0.55, N = 3SE +/- 0.13, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.71, N = 3SE +/- 0.05, N = 3SE +/- 0.35, N = 7SE +/- 0.43, N = 4SE +/- 0.15, N = 3SE +/- 0.22, N = 3SE +/- 0.07, N = 3SE +/- 0.41, N = 3SE +/- 0.33, N = 3SE +/- 0.18, N = 331.5329.9633.2930.9238.5235.6040.5235.1140.4535.0754.2538.4256.7742.9038.0036.4541.5734.8139.0233.5235.8835.791. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P816243240SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 3SE +/- 0.24, N = 3SE +/- 0.19, N = 3SE +/- 0.07, N = 3SE +/- 0.34, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.22, N = 3SE +/- 0.03, N = 3SE +/- 0.24, N = 7SE +/- 0.33, N = 4SE +/- 0.08, N = 3SE +/- 0.19, N = 3SE +/- 0.05, N = 3SE +/- 0.36, N = 3SE +/- 0.25, N = 3SE +/- 0.14, N = 331.7133.3830.0432.3425.9628.0924.6828.4824.7328.5218.4326.0317.6223.3126.3327.4524.0528.7325.6329.8427.8727.941. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 512 - Model: AlexNetXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P400800120016002000SE +/- 19.23, N = 3SE +/- 1.84, N = 3SE +/- 18.52, N = 3SE +/- 0.51, N = 3SE +/- 5.91, N = 3SE +/- 1.47, N = 3SE +/- 8.55, N = 3SE +/- 5.13, N = 3SE +/- 17.88, N = 4SE +/- 3.89, N = 3SE +/- 38.99, N = 12SE +/- 0.86, N = 3SE +/- 20.10, N = 12SE +/- 2.03, N = 3SE +/- 7.97, N = 3SE +/- 4.07, N = 3SE +/- 23.15, N = 4SE +/- 0.49, N = 3SE +/- 12.71, N = 3SE +/- 0.67, N = 3SE +/- 0.57, N = 3SE +/- 0.35, N = 32039.631486.121899.641278.451470.46883.001677.621108.951604.111047.961064.18732.79979.89677.401876.661620.531897.641392.911862.791363.33796.82875.23

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.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.32Configuration: Multi-ThreadedXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P160K320K480K640K800KSE +/- 248.00, N = 3SE +/- 63.14, N = 3SE +/- 362.38, N = 3SE +/- 105.60, N = 3SE +/- 457.00, N = 3SE +/- 395.04, N = 3SE +/- 160.59, N = 3SE +/- 40.48, N = 3SE +/- 286.86, N = 3SE +/- 155.19, N = 3SE +/- 1647.70, N = 3SE +/- 846.02, N = 3SE +/- 1610.78, N = 3SE +/- 306.14, N = 3SE +/- 804.57, N = 3SE +/- 156.35, N = 3SE +/- 1065.76, N = 3SE +/- 412.72, N = 3SE +/- 713.18, N = 3SE +/- 86.78, N = 3SE +/- 86.28, N = 3SE +/- 82.80, N = 3380867.5195648.8350187.5179647.5182366.991036.5316359.2162727.4288129.1143755.3519392.2262432.6456271.0233836.3759922.1374413.5644091.0319827.2593246.9294183.2153920.8172566.81. (CXX) g++ options: -O3 -march=native -fPIE -pie

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P160320480640800SE +/- 0.33, N = 3SE +/- 0.69, N = 3SE +/- 0.00, N = 3SE +/- 0.28, N = 3SE +/- 0.22, N = 3SE +/- 3.13, N = 3SE +/- 0.35, N = 3SE +/- 0.47, N = 3SE +/- 0.40, N = 3SE +/- 0.15, N = 3SE +/- 1.63, N = 3SE +/- 1.47, N = 3SE +/- 1.58, N = 3SE +/- 0.99, N = 3SE +/- 0.03, N = 3SE +/- 0.18, N = 3SE +/- 0.43, N = 3SE +/- 0.43, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.12, N = 3SE +/- 0.18, N = 3234.24224.64283.11356.38207.84415.89278.52257.50256.63236.52630.23618.97719.83692.79275.53551.07233.63471.98253.63510.10654.83592.221. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P120240360480600SE +/- 0.77, N = 3SE +/- 0.69, N = 3SE +/- 0.01, N = 3SE +/- 0.12, N = 3SE +/- 0.10, N = 3SE +/- 0.36, N = 3SE +/- 0.54, N = 3SE +/- 0.41, N = 3SE +/- 0.58, N = 3SE +/- 0.14, N = 3SE +/- 0.18, N = 3SE +/- 0.15, N = 3SE +/- 0.18, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.41, N = 3SE +/- 0.12, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3545.21284.64423.18167.8396.0847.99401.45217.11373.50202.59113.6858.0699.7251.85231.83115.78205.03101.51188.7493.9948.7353.891. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.16, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.23, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.14, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 348.7627.1042.8224.6422.5111.8639.3422.3635.9319.9123.1213.4021.2212.1763.8537.7460.3734.6452.6730.6116.9218.47

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Pabellon Barcelona - Compute: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.26, N = 3SE +/- 0.15, N = 3SE +/- 0.26, N = 3SE +/- 0.15, N = 3SE +/- 0.20, N = 3SE +/- 0.44, N = 3SE +/- 0.08, N = 3SE +/- 0.31, N = 3SE +/- 0.17, N = 3SE +/- 0.25, N = 3SE +/- 0.29, N = 3SE +/- 0.13, N = 3SE +/- 0.13, N = 3SE +/- 0.13, N = 3SE +/- 0.11, N = 3SE +/- 0.16, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.16, N = 341.8580.2946.3887.9276.10150.3951.2997.8055.26106.9631.2860.1135.1165.6721.6639.8323.8444.8726.3548.8894.4185.12

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.11, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 311.6911.1715.0416.509.8218.0814.4213.5913.3412.5424.2423.4926.1225.336.2711.354.918.485.209.4712.6711.221. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2K4K6K8K10KSE +/- 3.37, N = 3SE +/- 4.97, N = 3SE +/- 8.98, N = 3SE +/- 4.52, N = 3SE +/- 1.00, N = 3SE +/- 0.24, N = 3SE +/- 3.66, N = 3SE +/- 9.37, N = 3SE +/- 4.05, N = 3SE +/- 7.17, N = 3SE +/- 3.86, N = 3SE +/- 2.87, N = 3SE +/- 5.02, N = 3SE +/- 2.55, N = 3SE +/- 4.91, N = 3SE +/- 54.09, N = 3SE +/- 10.88, N = 3SE +/- 18.02, N = 3SE +/- 6.56, N = 3SE +/- 25.68, N = 3SE +/- 3.87, N = 3SE +/- 2.97, N = 310934.175725.037862.583606.122030.971100.187689.324093.737127.283802.612957.691525.262743.391414.8610182.165594.819752.895608.079203.195024.322506.102830.121. (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.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.12, N = 3SE +/- 0.57, N = 3SE +/- 0.11, N = 3SE +/- 0.18, N = 3SE +/- 0.56, N = 3SE +/- 0.49, N = 3SE +/- 0.23, N = 3SE +/- 0.31, N = 3SE +/- 0.50, N = 3SE +/- 0.23, N = 3SE +/- 0.21, N = 3SE +/- 0.31, N = 3SE +/- 0.90, N = 3SE +/- 0.10, N = 3SE +/- 0.30, N = 3SE +/- 0.60, N = 3SE +/- 0.60, N = 3SE +/- 0.57, N = 3SE +/- 0.65, N = 3SE +/- 0.60, N = 3SE +/- 1.29, N = 4SE +/- 1.08, N = 364.7950.8368.8362.1366.9159.9473.9763.5472.8061.25140.07105.22149.99116.64125.24103.60114.1681.43113.5281.60112.76102.051. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.02832, N = 3SE +/- 0.21736, N = 3SE +/- 0.02421, N = 3SE +/- 0.04566, N = 3SE +/- 0.12404, N = 3SE +/- 0.13407, N = 3SE +/- 0.04124, N = 3SE +/- 0.07567, N = 3SE +/- 0.09371, N = 3SE +/- 0.06023, N = 3SE +/- 0.01071, N = 3SE +/- 0.02804, N = 3SE +/- 0.03979, N = 3SE +/- 0.00728, N = 3SE +/- 0.01907, N = 3SE +/- 0.05587, N = 3SE +/- 0.04604, N = 3SE +/- 0.08493, N = 3SE +/- 0.05052, N = 3SE +/- 0.09004, N = 3SE +/- 0.10146, N = 4SE +/- 0.10485, N = 315.4341019.6773014.5280016.0956014.9460016.6863013.5190015.7394013.7366016.325107.139179.503536.667188.572867.984029.652968.7597112.280708.8093012.254908.871449.800901. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.16, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 36.956.727.729.7924.3923.257.977.767.327.1316.7316.3618.7517.814.378.033.786.933.987.239.468.591. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4K8K12K16K20KSE +/- 7.74, N = 3SE +/- 5.57, N = 3SE +/- 4.14, N = 3SE +/- 3.97, N = 3SE +/- 2.06, N = 3SE +/- 1.47, N = 3SE +/- 7.22, N = 3SE +/- 0.46, N = 3SE +/- 3.67, N = 3SE +/- 3.72, N = 3SE +/- 4.90, N = 3SE +/- 11.75, N = 3SE +/- 6.55, N = 3SE +/- 20.57, N = 3SE +/- 10.13, N = 3SE +/- 4.98, N = 3SE +/- 7.99, N = 3SE +/- 4.58, N = 3SE +/- 7.03, N = 3SE +/- 4.14, N = 3SE +/- 1.89, N = 3SE +/- 3.18, N = 318375.639515.8615519.925943.663263.031698.0514033.947207.7413102.286717.654259.992179.993787.942006.0014601.527862.6512665.626846.7912024.186554.123355.733692.671. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-50Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20406080100SE +/- 0.52, N = 3SE +/- 0.16, N = 3SE +/- 0.49, N = 3SE +/- 0.03, N = 3SE +/- 0.20, N = 3SE +/- 0.22, N = 3SE +/- 0.44, N = 3SE +/- 0.36, N = 3SE +/- 0.38, N = 8SE +/- 0.28, N = 3SE +/- 0.22, N = 4SE +/- 0.45, N = 4SE +/- 0.20, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.37, N = 12SE +/- 0.15, N = 3SE +/- 0.28, N = 15SE +/- 0.15, N = 3SE +/- 0.30, N = 3SE +/- 0.12, N = 345.6977.5235.5351.4642.6461.3235.4648.5946.4950.9917.9135.8516.4532.4514.9031.6219.8441.1719.3440.4540.9741.11

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P714212835SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 4SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.482.362.503.588.278.582.792.672.532.5024.7724.2528.2427.0111.2010.929.259.4210.1710.0913.1111.781. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P10K20K30K40K50KSE +/- 39.08, N = 3SE +/- 47.41, N = 3SE +/- 53.98, N = 3SE +/- 2.71, N = 3SE +/- 6.41, N = 3SE +/- 5.62, N = 3SE +/- 233.95, N = 3SE +/- 21.27, N = 3SE +/- 80.58, N = 3SE +/- 234.43, N = 4SE +/- 19.55, N = 3SE +/- 6.61, N = 3SE +/- 12.93, N = 3SE +/- 7.59, N = 3SE +/- 11.27, N = 3SE +/- 5.50, N = 3SE +/- 110.20, N = 3SE +/- 11.71, N = 3SE +/- 147.69, N = 3SE +/- 4.46, N = 3SE +/- 1.69, N = 3SE +/- 0.41, N = 348122.1626943.4247654.9515943.179657.084502.7440032.8120910.5337710.6019183.8311390.045866.6310020.915278.4622499.0311508.1920550.939997.8918666.169356.714830.015366.791. (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.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P48121620SE +/- 0.02114, N = 3SE +/- 0.00434, N = 3SE +/- 0.03717, N = 3SE +/- 0.01428, N = 3SE +/- 0.06784, N = 3SE +/- 0.01980, N = 3SE +/- 0.01417, N = 3SE +/- 0.07483, N = 3SE +/- 0.08283, N = 3SE +/- 0.02823, N = 3SE +/- 0.01174, N = 3SE +/- 0.01181, N = 3SE +/- 0.11771, N = 3SE +/- 0.00515, N = 3SE +/- 0.01997, N = 3SE +/- 0.00921, N = 3SE +/- 0.03939, N = 3SE +/- 0.02436, N = 3SE +/- 0.02885, N = 3SE +/- 0.00351, N = 3SE +/- 0.02580, N = 3SE +/- 0.03809, N = 35.109484.016876.906195.215206.507475.543797.308225.290277.348485.2278116.541807.1353517.055307.5958910.066607.280879.066076.184789.124526.377807.705117.745211. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50100150200250SE +/- 0.81, N = 3SE +/- 0.27, N = 3SE +/- 0.78, N = 3SE +/- 0.53, N = 3SE +/- 1.59, N = 3SE +/- 0.64, N = 3SE +/- 0.26, N = 3SE +/- 2.63, N = 3SE +/- 1.53, N = 3SE +/- 1.04, N = 3SE +/- 0.04, N = 3SE +/- 0.23, N = 3SE +/- 0.40, N = 3SE +/- 0.09, N = 3SE +/- 0.19, N = 3SE +/- 0.17, N = 3SE +/- 0.48, N = 3SE +/- 0.64, N = 3SE +/- 0.35, N = 3SE +/- 0.08, N = 3SE +/- 0.43, N = 3SE +/- 0.64, N = 3195.45248.68144.64191.56153.54180.23136.69188.93135.98191.1160.39140.0358.58131.5499.22137.19110.17161.50109.46156.61129.65128.981. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 35.185.195.726.435.936.566.025.775.555.2716.9816.4119.2318.206.606.255.345.415.985.857.606.771. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P8K16K24K32K40KSE +/- 17.69, N = 3SE +/- 4.30, N = 3SE +/- 12.64, N = 3SE +/- 5.45, N = 3SE +/- 16.45, N = 3SE +/- 15.90, N = 3SE +/- 3.89, N = 3SE +/- 7.39, N = 3SE +/- 12.64, N = 3SE +/- 18.13, N = 3SE +/- 41.98, N = 3SE +/- 0.67, N = 3SE +/- 30.29, N = 3SE +/- 2.54, N = 3SE +/- 10.72, N = 3SE +/- 8.57, N = 3SE +/- 15.46, N = 3SE +/- 7.70, N = 3SE +/- 46.87, N = 3SE +/- 2.59, N = 3SE +/- 7.49, N = 3SE +/- 2.45, N = 324676.8612317.9420937.589004.2313467.285913.0418561.919686.3917252.279087.1216394.068536.3914551.127721.0038546.6619917.5435809.5917230.3531974.8816038.908279.639256.111. (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.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: yolov4 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50100150200250SE +/- 0.57, N = 3SE +/- 0.30, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.46, N = 3SE +/- 0.36, N = 3SE +/- 0.13, N = 3SE +/- 0.01, N = 3SE +/- 0.74, N = 3SE +/- 0.22, N = 3SE +/- 1.41, N = 3SE +/- 0.52, N = 3SE +/- 2.49, N = 3SE +/- 0.91, N = 3SE +/- 2.38, N = 3SE +/- 0.20, N = 3SE +/- 0.15, N = 3SE +/- 0.05, N = 3SE +/- 0.61, N = 3SE +/- 0.20, N = 3SE +/- 0.53, N = 3SE +/- 0.01, N = 370.4447.9078.4055.1270.3456.5879.3861.0176.4261.83219.17129.94217.93140.14179.99115.79158.27102.04157.51102.19118.02114.171. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: yolov4 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.11392, N = 3SE +/- 0.13074, N = 3SE +/- 0.01439, N = 3SE +/- 0.01672, N = 3SE +/- 0.09305, N = 3SE +/- 0.11211, N = 3SE +/- 0.02010, N = 3SE +/- 0.00279, N = 3SE +/- 0.12795, N = 3SE +/- 0.05802, N = 3SE +/- 0.02942, N = 3SE +/- 0.03075, N = 3SE +/- 0.05214, N = 3SE +/- 0.04609, N = 3SE +/- 0.07368, N = 3SE +/- 0.01474, N = 3SE +/- 0.00606, N = 3SE +/- 0.00491, N = 3SE +/- 0.02467, N = 3SE +/- 0.01934, N = 3SE +/- 0.03807, N = 3SE +/- 0.00112, N = 314.1980020.8795012.7546018.1413014.2169017.6747012.5974016.3898013.0869016.174204.562787.695834.589607.135875.557678.635946.318289.800086.348919.785168.472838.758141. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Noise-GaussianXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.88, N = 3SE +/- 0.00, N = 3SE +/- 1.15, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 1.44, N = 4SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 0.67, N = 3SE +/- 0.00, N = 3SE +/- 0.67, N = 3SE +/- 0.58, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 31941711821581531211641581501501261551181451641511791651781601201231. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 23.6Input: Carbon NanotubeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20406080100SE +/- 0.10, N = 3SE +/- 0.12, N = 3SE +/- 0.55, N = 3SE +/- 0.24, N = 3SE +/- 0.60, N = 3SE +/- 0.12, N = 3SE +/- 0.27, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.18, N = 3SE +/- 0.14, N = 3SE +/- 0.30, N = 15SE +/- 0.11, N = 3SE +/- 0.29, N = 15SE +/- 0.21, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 338.0543.0541.5647.7848.4576.1843.9148.4942.8754.3827.1533.1224.1330.0025.8434.8766.7163.801. (CC) gcc options: -shared -lxc -lblas -lmpi

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.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: SharpenXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P80160240320400SE +/- 1.67, N = 3SE +/- 0.00, N = 3SE +/- 1.67, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 1.20, N = 3SE +/- 0.00, N = 3SE +/- 0.67, N = 3SE +/- 0.00, N = 3SE +/- 0.88, N = 3SE +/- 0.00, N = 3SE +/- 3.00, N = 3SE +/- 0.33, N = 3SE +/- 0.88, N = 3SE +/- 0.00, N = 3SE +/- 1.20, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3284167265157144792271301991143812813362533102072972012621731051121. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: defconfigXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1326395265SE +/- 0.23, N = 7SE +/- 0.39, N = 4SE +/- 0.43, N = 3SE +/- 0.55, N = 3SE +/- 0.42, N = 5SE +/- 0.33, N = 3SE +/- 0.31, N = 5SE +/- 0.57, N = 3SE +/- 0.40, N = 4SE +/- 0.59, N = 3SE +/- 0.28, N = 8SE +/- 0.37, N = 6SE +/- 0.22, N = 14SE +/- 0.39, N = 6SE +/- 0.28, N = 7SE +/- 0.44, N = 3SE +/- 0.17, N = 13SE +/- 0.46, N = 3SE +/- 0.33, N = 3SE +/- 0.43, N = 3SE +/- 0.54, N = 4SE +/- 0.50, N = 425.8734.5929.9839.4539.4155.7430.5340.5032.1443.4232.6637.6435.5541.1030.6838.5325.8732.8927.6834.5746.3544.28

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.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: EnhancedXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P90180270360450SE +/- 1.15, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 1.45, N = 3SE +/- 0.00, N = 3SE +/- 1.00, N = 3SE +/- 0.00, N = 3SE +/- 2.60, N = 3SE +/- 0.00, N = 3SE +/- 1.53, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 1.20, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 33742283422162151233011812691603892993502704033044083003682641631721. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P48121620SE +/- 0.02988, N = 3SE +/- 0.03569, N = 3SE +/- 0.02062, N = 3SE +/- 0.05586, N = 3SE +/- 0.02822, N = 3SE +/- 0.01290, N = 3SE +/- 0.02155, N = 3SE +/- 0.01694, N = 3SE +/- 0.03854, N = 3SE +/- 0.02534, N = 3SE +/- 0.02665, N = 3SE +/- 0.00657, N = 3SE +/- 0.06373, N = 3SE +/- 0.00867, N = 3SE +/- 0.13507, N = 3SE +/- 0.02898, N = 3SE +/- 0.05412, N = 3SE +/- 0.08338, N = 3SE +/- 0.06993, N = 3SE +/- 0.03015, N = 3SE +/- 0.07856, N = 3SE +/- 0.05765, N = 34.654314.263525.292874.866045.597635.671755.886285.260755.749215.1091415.2784013.1789016.5828014.4761012.8093010.5537011.458009.1852411.502409.0895510.3977010.302601. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: ParallelXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50100150200250SE +/- 1.38, N = 3SE +/- 1.98, N = 3SE +/- 0.74, N = 3SE +/- 2.34, N = 3SE +/- 0.90, N = 3SE +/- 0.40, N = 3SE +/- 0.62, N = 3SE +/- 0.61, N = 3SE +/- 1.17, N = 3SE +/- 0.98, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.23, N = 3SE +/- 0.04, N = 3SE +/- 0.83, N = 3SE +/- 0.26, N = 3SE +/- 0.41, N = 3SE +/- 0.99, N = 3SE +/- 0.53, N = 3SE +/- 0.36, N = 3SE +/- 0.72, N = 3SE +/- 0.54, N = 3214.81234.54188.89205.53178.62176.29169.85190.06173.91195.7165.4375.8760.2969.0778.0794.7487.26108.8786.92110.0096.1797.051. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: SwirlXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P150300450600750SE +/- 0.58, N = 3SE +/- 0.88, N = 3SE +/- 1.76, N = 3SE +/- 1.15, N = 3SE +/- 4.41, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.67, N = 3SE +/- 1.33, N = 3SE +/- 0.58, N = 3SE +/- 3.00, N = 3SE +/- 1.15, N = 3SE +/- 2.33, N = 3SE +/- 4.91, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 36895296175024032785584325093945175664745045675355835465535113633791. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Random ReadXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P300M600M900M1200M1500MSE +/- 82996.27, N = 3SE +/- 1119855.47, N = 3SE +/- 1411218.60, N = 3SE +/- 3338569.32, N = 3SE +/- 704022.38, N = 3SE +/- 744772.41, N = 3SE +/- 785893.13, N = 3SE +/- 86315.43, N = 3SE +/- 1120857.00, N = 3SE +/- 81327.78, N = 3SE +/- 1897433.73, N = 3SE +/- 2130676.22, N = 3SE +/- 4448269.22, N = 3SE +/- 1336933.52, N = 3SE +/- 1154820.73, N = 3SE +/- 153963.59, N = 3SE +/- 2927744.32, N = 3SE +/- 326038.01, N = 3SE +/- 599126.55, N = 3SE +/- 610621.13, N = 3SE +/- 211259.57, N = 3SE +/- 206014.27, N = 357839280029751794652504999826536617025123410312558495347375594623446993940553317320534615412729075616453094831115720742574749552113520656457323545210312091515233100918151451104092729492349926362661865371. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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.

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P70140210280350SE +/- 1.93, N = 15SE +/- 0.03, N = 3SE +/- 0.99, N = 3SE +/- 0.31, N = 3SE +/- 0.75, N = 3SE +/- 0.11, N = 3SE +/- 0.93, N = 3SE +/- 0.96, N = 4SE +/- 1.48, N = 3SE +/- 0.31, N = 3SE +/- 0.08, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 3.85, N = 3SE +/- 0.01, N = 3SE +/- 2.28, N = 3SE +/- 1.71, N = 3SE +/- 0.08, N = 3SE +/- 0.93, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3191.82108.12140.8386.9797.0849.20133.9978.11127.8570.2186.8843.7378.2439.35288.54232.75309.22187.68291.28171.7797.98107.721. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P17003400510068008500SE +/- 48.20, N = 15SE +/- 0.70, N = 3SE +/- 24.63, N = 3SE +/- 7.77, N = 3SE +/- 18.66, N = 3SE +/- 2.69, N = 3SE +/- 23.14, N = 3SE +/- 23.93, N = 4SE +/- 36.96, N = 3SE +/- 7.81, N = 3SE +/- 1.96, N = 3SE +/- 0.09, N = 3SE +/- 0.82, N = 3SE +/- 0.03, N = 3SE +/- 96.25, N = 3SE +/- 0.20, N = 3SE +/- 57.10, N = 3SE +/- 42.85, N = 3SE +/- 2.03, N = 3SE +/- 23.15, N = 3SE +/- 0.10, N = 3SE +/- 0.13, N = 34795.462703.093520.712174.192426.881230.083349.761952.613196.121755.332172.101093.211955.88983.717213.435818.747730.474692.117281.964294.282449.452692.871. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2M4M6M8M10MSE +/- 9700.21, N = 3SE +/- 4635.95, N = 3SE +/- 12743.41, N = 3SE +/- 9256.47, N = 3SE +/- 30163.42, N = 15SE +/- 927.64, N = 3SE +/- 20649.43, N = 3SE +/- 5202.86, N = 3SE +/- 7219.73, N = 3SE +/- 1747.99, N = 3SE +/- 7619.10, N = 3SE +/- 40214.40, N = 15SE +/- 9906.84, N = 3SE +/- 10294.77, N = 3SE +/- 54771.53, N = 3SE +/- 4775.53, N = 3SE +/- 76743.24, N = 3SE +/- 54326.71, N = 3SE +/- 2306.54, N = 3SE +/- 10132.66, N = 3SE +/- 11678.22, N = 3SE +/- 1374.66, N = 34966250.782369998.363902076.242194630.601839444.401218453.343279262.531855135.332993136.601614763.707302247.143340274.036575205.542915609.758171241.714525288.008448083.734152864.887975907.453926265.282069432.532274636.131. (CC) gcc options: -O2 -lrt" -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.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput TotalXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P7K14K21K28K35KSE +/- 356.81, N = 15SE +/- 298.34, N = 15SE +/- 312.17, N = 3SE +/- 236.27, N = 4SE +/- 308.01, N = 15SE +/- 44.42, N = 3SE +/- 490.54, N = 15SE +/- 281.72, N = 15SE +/- 175.78, N = 4SE +/- 938.95, N = 15SE +/- 242.64, N = 3SE +/- 134.28, N = 4SE +/- 319.61, N = 3SE +/- 150.02, N = 4SE +/- 2115.57, N = 15SE +/- 237.63, N = 3SE +/- 343.97, N = 3SE +/- 348.28, N = 4SE +/- 526.01, N = 15SE +/- 357.13, N = 12SE +/- 58.21, N = 3SE +/- 207.19, N = 530547.626378.130108.724688.518943.613537.711866.715081.030236.021286.633010.529804.431677.227281.929877.125194.625977.429384.922931.827695.219485.819375.91. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Classroom - Compute: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.22, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.16, N = 3SE +/- 0.38, N = 3SE +/- 0.25, N = 3SE +/- 0.40, N = 3SE +/- 0.38, N = 3SE +/- 0.20, N = 3SE +/- 0.05, N = 3SE +/- 0.67, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.14, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 3SE +/- 0.18, N = 3SE +/- 0.04, N = 333.6163.6135.9168.6162.54124.6340.3876.8744.7186.6724.6148.0727.4152.4716.8231.6518.6735.6621.1240.2675.2667.67

Aircrack-ng

Aircrack-ng is a tool for assessing WiFi/WLAN network security. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgk/s, More Is BetterAircrack-ng 1.7Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780EXeon 6766EEPYC 9754EPYC 9684XEPYC 9654EPYC 8534PNEPYC 8534P40K80K120K160K200KSE +/- 773.70, N = 3SE +/- 840.59, N = 3SE +/- 1197.41, N = 15SE +/- 1953.08, N = 3SE +/- 1234.47, N = 15SE +/- 728.55, N = 3SE +/- 1131.21, N = 15SE +/- 122.31, N = 3SE +/- 1027.07, N = 15SE +/- 36.42, N = 3SE +/- 888.98, N = 3SE +/- 1379.58, N = 4SE +/- 714.86, N = 3SE +/- 1622.56, N = 3SE +/- 207.11, N = 3SE +/- 319.60, N = 3SE +/- 338.71, N = 364181.29153584.2053914.93147469.33159709.75101665.4765663.82134048.9683665.76117588.99116716.40114715.79177989.89206619.29186675.43141831.78153818.031. (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

EPYC 9684X 2P: The test run did not produce a result.

EPYC 9654 2P: The test run did not produce a result.

EPYC 9754 2P: The test run did not produce a result.

Xeon 6766E 2P: The test run did not produce a result.

Xeon 6780E 2P: The test run did not produce a result.

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.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Lossless, Highest CompressionXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.1350.270.4050.540.675SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.590.600.510.520.490.500.540.550.540.550.380.390.350.350.480.480.560.570.570.570.480.481. (CC) gcc options: -fvisibility=hidden -O2 -lm

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 2Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20406080100SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.16, N = 3SE +/- 0.04, N = 3SE +/- 0.25, N = 3SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.31, N = 3SE +/- 0.22, N = 3SE +/- 0.21, N = 3SE +/- 0.09, N = 3SE +/- 0.41, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.43, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 333.6534.7937.3137.7547.5152.2338.1439.8838.6541.1371.9769.3479.2076.4741.9240.3535.3234.0435.4034.1542.6141.661. (CXX) g++ options: -O3 -fPIC -lm

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.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput TotalXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2K4K6K8K10KSE +/- 0.12, N = 3SE +/- 18.09, N = 3SE +/- 47.68, N = 6SE +/- 18.43, N = 3SE +/- 0.17, N = 3SE +/- 0.07, N = 3SE +/- 0.15, N = 3SE +/- 18.30, N = 3SE +/- 26.47, N = 3SE +/- 0.09, N = 3SE +/- 14.40, N = 3SE +/- 0.00, N = 3SE +/- 36.60, N = 3SE +/- 8.17, N = 3SE +/- 76.45, N = 3SE +/- 63.64, N = 5SE +/- 2.29, N = 3SE +/- 55.17, N = 3SE +/- 74.95, N = 4SE +/- 0.20, N = 3SE +/- 19.02, N = 13SE +/- 25.25, N = 35493.43088.74853.62616.93288.41770.04386.62514.03787.02170.73414.22118.63067.01953.18396.75973.86999.75656.96835.75239.62818.53193.01. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

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.

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.7Test: llava-v1.5-7b-q4 - Acceleration: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 1.08, N = 12SE +/- 0.17, N = 4SE +/- 0.09, N = 15SE +/- 0.15, N = 4SE +/- 0.10, N = 15SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.30, N = 15SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 4SE +/- 0.05, N = 4SE +/- 0.02, N = 4SE +/- 0.10, N = 5SE +/- 0.07, N = 4SE +/- 0.02, N = 5SE +/- 0.07, N = 4SE +/- 0.11, N = 47.4916.6110.4420.9512.2919.117.268.137.777.672.999.602.838.5610.9420.6414.2925.3814.4725.7520.6521.00

LULESH

LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P15K30K45K60K75KSE +/- 359.29, N = 3SE +/- 437.51, N = 4SE +/- 272.44, N = 4SE +/- 182.38, N = 15SE +/- 188.07, N = 8SE +/- 53.94, N = 5SE +/- 207.54, N = 5SE +/- 217.67, N = 6SE +/- 206.20, N = 5SE +/- 211.12, N = 6SE +/- 91.45, N = 3SE +/- 314.85, N = 4SE +/- 55.60, N = 3SE +/- 254.25, N = 6SE +/- 427.75, N = 3SE +/- 244.17, N = 3SE +/- 675.87, N = 3SE +/- 194.65, N = 15SE +/- 349.39, N = 15SE +/- 372.12, N = 3SE +/- 373.78, N = 12SE +/- 372.63, N = 1270826.7338433.9743724.0623072.3021389.8217348.4353269.4725831.8452102.6723848.5444136.5326864.1142946.2626024.6046126.0824218.8354678.4927287.2648760.8226082.7116855.8916949.591. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

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.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression RatingXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P300K600K900K1200K1500KSE +/- 28320.74, N = 3SE +/- 119.21, N = 3SE +/- 185.14, N = 3SE +/- 61.74, N = 3SE +/- 559.82, N = 3SE +/- 309.57, N = 3SE +/- 6276.88, N = 3SE +/- 374.59, N = 3SE +/- 2598.93, N = 3SE +/- 916.22, N = 3SE +/- 903.01, N = 3SE +/- 258.19, N = 3SE +/- 2333.15, N = 3SE +/- 343.82, N = 3SE +/- 14188.09, N = 3SE +/- 291.68, N = 3SE +/- 38431.24, N = 3SE +/- 1085.99, N = 3SE +/- 15794.26, N = 3SE +/- 111.53, N = 3SE +/- 81.21, N = 3SE +/- 120.28, N = 35734103381465570542878623665341883454691312433474017072077388463354637687673134210631340226780871111646571606510332776426263383553679891. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression RatingXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P170K340K510K680K850KSE +/- 709.87, N = 3SE +/- 1982.97, N = 3SE +/- 4200.62, N = 3SE +/- 1535.53, N = 3SE +/- 1785.84, N = 3SE +/- 248.73, N = 3SE +/- 3924.87, N = 3SE +/- 2258.81, N = 3SE +/- 5127.15, N = 3SE +/- 1836.55, N = 3SE +/- 1102.28, N = 3SE +/- 933.46, N = 3SE +/- 7384.79, N = 3SE +/- 610.46, N = 3SE +/- 2500.52, N = 3SE +/- 1188.96, N = 3SE +/- 3075.95, N = 3SE +/- 1520.59, N = 3SE +/- 6869.20, N = 3SE +/- 719.89, N = 3SE +/- 206.69, N = 3SE +/- 268.94, N = 36742394416365927293881903917192437325255143423634917293147166112004380485918354138517769685754407413665791127617345495003190943313571. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

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.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1000M2000M3000M4000M5000MSE +/- 9014927.20, N = 3SE +/- 4000850.45, N = 3SE +/- 2438223.49, N = 3SE +/- 468935.32, N = 3SE +/- 467864.03, N = 3SE +/- 971797.70, N = 3SE +/- 9164253.67, N = 3SE +/- 1778769.08, N = 3SE +/- 7233468.75, N = 3SE +/- 3359086.30, N = 3SE +/- 235140242.43, N = 12SE +/- 433397.56, N = 3SE +/- 6340323.98, N = 3SE +/- 4963892.66, N = 3SE +/- 8808919.44, N = 3SE +/- 1066998.75, N = 3SE +/- 291866.98, N = 3SE +/- 2902028.33, N = 3360964633318147600003124402667160509700072315353310736276673205161000159799333331562616671588496667278303908322753526674467281333237614900043249303332282802000114710800011500526671. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

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.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 512Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4M8M12M16M20MSE +/- 162096.41, N = 15SE +/- 9492.69, N = 3SE +/- 38670.98, N = 3SE +/- 8685.88, N = 3SE +/- 11150.49, N = 3SE +/- 18478.22, N = 3SE +/- 7571.88, N = 3SE +/- 30179.10, N = 3SE +/- 17457.89, N = 3SE +/- 5238.74, N = 3SE +/- 133.33, N = 3SE +/- 88.19, N = 3SE +/- 33.33, N = 3SE +/- 115.47, N = 3SE +/- 11050.39, N = 3SE +/- 120841.84, N = 4SE +/- 234395.65, N = 15SE +/- 134727.30, N = 3SE +/- 124115.99, N = 3SE +/- 121304.48, N = 3SE +/- 31571.79, N = 3SE +/- 3464.10, N = 318200533187743331684333316882333135990001360833316884000168853331690466716906333696996769839336280767628430011273667111745001308313313387333133906671339466711236333112900001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P5K10K15K20K25KSE +/- 30.19, N = 4SE +/- 22.33, N = 3SE +/- 100.23, N = 4SE +/- 6.77, N = 3SE +/- 17.53, N = 3SE +/- 13.82, N = 3SE +/- 10.93, N = 4SE +/- 6.53, N = 3SE +/- 4.83, N = 4SE +/- 155.10, N = 12SE +/- 1274.85, N = 12SE +/- 367.88, N = 12SE +/- 602.37, N = 15SE +/- 309.03, N = 15SE +/- 530.31, N = 15SE +/- 122.47, N = 15SE +/- 50.01, N = 8SE +/- 14.68, N = 316004.078089.1414946.767562.808880.824479.2812793.016445.0910970.455366.6922017.7512575.3022166.6011424.0519414.9410652.035745.046416.341. (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

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.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: Very ThoroughXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.0007, N = 3SE +/- 0.0294, N = 3SE +/- 0.0379, N = 3SE +/- 0.0054, N = 3SE +/- 0.0026, N = 3SE +/- 0.0075, N = 3SE +/- 0.0022, N = 3SE +/- 0.0016, N = 3SE +/- 0.0055, N = 3SE +/- 0.0016, N = 3SE +/- 0.0314, N = 3SE +/- 0.0005, N = 3SE +/- 0.0386, N = 3SE +/- 0.0005, N = 3SE +/- 0.1961, N = 15SE +/- 0.0006, N = 3SE +/- 0.1284, N = 3SE +/- 0.0254, N = 3SE +/- 0.1965, N = 3SE +/- 0.0006, N = 3SE +/- 0.0012, N = 3SE +/- 0.0007, N = 314.63347.624013.75187.08854.29352.159611.88755.970210.20515.134212.62206.516611.35735.863222.002312.318719.405210.322417.00989.18445.24065.74201. (CXX) g++ options: -O3 -flto -pthread

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 256 - Model: AlexNetXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P400800120016002000SE +/- 1.86, N = 3SE +/- 4.27, N = 3SE +/- 5.06, N = 3SE +/- 0.54, N = 3SE +/- 3.03, N = 3SE +/- 1.28, N = 3SE +/- 5.81, N = 3SE +/- 6.82, N = 3SE +/- 1.15, N = 3SE +/- 1.21, N = 3SE +/- 10.51, N = 4SE +/- 2.12, N = 3SE +/- 19.06, N = 15SE +/- 1.03, N = 3SE +/- 10.23, N = 15SE +/- 4.01, N = 3SE +/- 15.82, N = 3SE +/- 1.54, N = 3SE +/- 6.01, N = 3SE +/- 2.58, N = 3SE +/- 0.41, N = 3SE +/- 0.99, N = 31681.441277.461730.251205.301340.34852.141528.411058.991479.53986.041010.93668.33857.20609.671403.461454.711560.561301.261524.741278.58771.47843.63

miniFE

MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P13K26K39K52K65KSE +/- 1633.92, N = 15SE +/- 135.77, N = 4SE +/- 467.78, N = 4SE +/- 11.14, N = 4SE +/- 366.08, N = 4SE +/- 54.57, N = 4SE +/- 283.44, N = 3SE +/- 572.00, N = 12SE +/- 1097.75, N = 15SE +/- 601.03, N = 12SE +/- 712.94, N = 12SE +/- 8.42, N = 4SE +/- 1039.71, N = 15SE +/- 19.87, N = 4SE +/- 2087.13, N = 15SE +/- 53.07, N = 5SE +/- 2182.93, N = 15SE +/- 92.04, N = 5SE +/- 1510.37, N = 15SE +/- 42.89, N = 5SE +/- 42.17, N = 4SE +/- 49.20, N = 442114.631830.246877.535230.131046.522957.321583.218621.219876.520091.137461.140626.439580.940441.042113.550139.959466.653167.843531.050909.624548.824526.01. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Cell Phone Drop TestXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1020304050SE +/- 0.32, N = 3SE +/- 0.18, N = 3SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.21, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.26, N = 3SE +/- 0.05, N = 3SE +/- 0.19, N = 3SE +/- 0.05, N = 3SE +/- 0.26, N = 3SE +/- 0.35, N = 3SE +/- 0.18, N = 326.8827.9028.2231.4931.1743.4527.6331.9227.5133.7527.8623.6922.5220.4523.5322.5335.9533.23

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.11Video Input: Bosphorus 4K - Video Preset: FasterXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780EXeon 6766EEPYC 9754EPYC 9684XEPYC 9654EPYC 8534PNEPYC 8534P48121620SE +/- 0.12, N = 3SE +/- 0.08, N = 3SE +/- 0.16, N = 3SE +/- 0.02, N = 3SE +/- 0.13, N = 6SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.13, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 313.2615.9712.5415.0613.0213.1011.9714.3611.9014.1011.1810.3914.7917.0216.7313.6113.971. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

Video Input: Bosphorus 4K - Video Preset: Faster

EPYC 9684X 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1

EPYC 9654 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

Xeon 6766E 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1

Xeon 6780E 2P: The test quit with a non-zero exit status. E: Parameter Check Error: Number of threads out of range (-1

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e13Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20406080100SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.25, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 4SE +/- 0.02, N = 3SE +/- 0.02, N = 4SE +/- 0.02, N = 3SE +/- 0.01, N = 4SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 325.0248.4127.1151.8240.4481.3630.2058.1633.9267.3226.4652.0729.3957.7111.5621.7612.1323.2712.6324.2454.3945.951. (CXX) g++ options: -O3

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.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2024Implementation: MPI CPU - Input: water_GMX50_bareXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.046, N = 3SE +/- 0.009, N = 3SE +/- 0.019, N = 3SE +/- 0.010, N = 3SE +/- 0.036, N = 3SE +/- 0.018, N = 3SE +/- 0.032, N = 3SE +/- 0.007, N = 3SE +/- 0.025, N = 3SE +/- 0.019, N = 3SE +/- 0.067, N = 3SE +/- 0.032, N = 3SE +/- 0.224, N = 14SE +/- 0.052, N = 3SE +/- 0.189, N = 6SE +/- 0.034, N = 3SE +/- 0.004, N = 3SE +/- 0.004, N = 318.54010.47715.4508.7628.9644.81314.1278.28213.3927.54521.30513.24222.52515.22220.23412.5146.2466.4991. (CXX) g++ options: -O3 -lm

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.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ExhaustiveXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P48121620SE +/- 0.0157, N = 3SE +/- 0.0072, N = 3SE +/- 0.0164, N = 3SE +/- 0.0024, N = 3SE +/- 0.0029, N = 3SE +/- 0.0016, N = 3SE +/- 0.0016, N = 3SE +/- 0.0011, N = 3SE +/- 0.0033, N = 3SE +/- 0.0003, N = 3SE +/- 0.0118, N = 3SE +/- 0.0000, N = 3SE +/- 0.0040, N = 3SE +/- 0.0007, N = 3SE +/- 0.1160, N = 3SE +/- 0.0025, N = 3SE +/- 0.0170, N = 3SE +/- 0.0086, N = 3SE +/- 0.0218, N = 3SE +/- 0.0002, N = 3SE +/- 0.0006, N = 3SE +/- 0.0010, N = 38.90914.63038.37014.26892.64711.33197.20753.62766.21263.12047.90604.05747.12913.650514.10387.557311.91076.302710.60185.63013.19933.51461. (CXX) g++ options: -O3 -flto -pthread

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.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: MediumXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1020304050SE +/- 0.05, N = 3SE +/- 0.16, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.16, N = 15SE +/- 0.01, N = 3SE +/- 0.23, N = 5SE +/- 0.01, N = 3SE +/- 0.28, N = 11SE +/- 0.05, N = 3SE +/- 0.46, N = 4SE +/- 0.03, N = 3SE +/- 0.23, N = 4SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 330.3716.6427.6517.8516.1912.5624.6515.6822.2515.1722.5917.2922.0315.6538.1232.4445.2632.7442.0331.4417.8019.25

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSKXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P300K600K900K1200K1500KSE +/- 15579.94, N = 15SE +/- 864.17, N = 3SE +/- 2519.00, N = 3SE +/- 1599.31, N = 3SE +/- 1491.72, N = 3SE +/- 177.03, N = 3SE +/- 788.50, N = 3SE +/- 1274.00, N = 3SE +/- 5122.51, N = 3SE +/- 987.33, N = 3SE +/- 394.40, N = 3SE +/- 66.16, N = 3SE +/- 187.67, N = 3SE +/- 30.00, N = 3SE +/- 4358.90, N = 3SE +/- 478.70, N = 3SE +/- 5291.50, N = 3SE +/- 2683.36, N = 3SE +/- 1333.33, N = 3SE +/- 921.74, N = 3SE +/- 79.20, N = 3SE +/- 321.00, N = 37970934281167932534012784915832480666578893340765628752865524072562098033676571889251551000803187129400067132212533336434753399983862121. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 1 - Model: GoogLeNetXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1224364860SE +/- 0.12, N = 6SE +/- 0.21, N = 6SE +/- 0.23, N = 15SE +/- 0.19, N = 6SE +/- 0.34, N = 15SE +/- 0.24, N = 7SE +/- 0.23, N = 12SE +/- 0.27, N = 6SE +/- 0.20, N = 15SE +/- 0.27, N = 7SE +/- 0.06, N = 9SE +/- 0.09, N = 4SE +/- 0.05, N = 7SE +/- 0.04, N = 4SE +/- 0.04, N = 3SE +/- 0.27, N = 15SE +/- 0.17, N = 15SE +/- 0.33, N = 15SE +/- 0.14, N = 15SE +/- 0.41, N = 15SE +/- 0.14, N = 6SE +/- 0.30, N = 1512.6921.6713.5033.5629.1853.3814.6731.4718.2436.136.7213.145.8812.408.1021.0012.8334.2012.4232.8036.4836.66

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.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 57Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P16M32M48M64M80MSE +/- 68119.99, N = 3SE +/- 599433.09, N = 9SE +/- 3844.19, N = 3SE +/- 666.67, N = 3SE +/- 178712.37, N = 3SE +/- 15982.63, N = 3SE +/- 822374.71, N = 4SE +/- 5364.49, N = 3SE +/- 4163.33, N = 3SE +/- 4910.31, N = 3SE +/- 11392.00, N = 3SE +/- 5206.83, N = 3SE +/- 2185.81, N = 3SE +/- 20835.33, N = 3SE +/- 8089.77, N = 3SE +/- 7937.25, N = 3SE +/- 34724.31, N = 3SE +/- 7423.69, N = 3SE +/- 486361.09, N = 3SE +/- 12719.19, N = 3SE +/- 393224.42, N = 3SE +/- 4807.40, N = 3704990007314066766991333670746675393033354089667662660006712766767064000671303333712133337195667334393333344466746229667462700005541566755455333549546675542533345486333462106671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.32Configuration: Single-ThreadedXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P8001600240032004000SE +/- 30.36, N = 12SE +/- 33.58, N = 3SE +/- 25.40, N = 3SE +/- 28.42, N = 3SE +/- 2.82, N = 3SE +/- 14.88, N = 3SE +/- 28.57, N = 3SE +/- 25.35, N = 3SE +/- 28.11, N = 3SE +/- 22.17, N = 3SE +/- 7.35, N = 3SE +/- 6.88, N = 3SE +/- 4.34, N = 3SE +/- 4.87, N = 3SE +/- 19.30, N = 3SE +/- 14.83, N = 3SE +/- 1.14, N = 3SE +/- 22.39, N = 3SE +/- 19.83, N = 3SE +/- 24.27, N = 3SE +/- 14.40, N = 3SE +/- 18.64, N = 33586.33600.53210.13209.62558.62534.53222.83229.13207.03225.71954.21960.71760.81765.62590.82596.13125.43114.23098.53105.92593.02593.31. (CXX) g++ options: -O3 -march=native -fPIE -pie

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.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 512Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P200M400M600M800M1000MSE +/- 7562646.32, N = 8SE +/- 1515552.70, N = 3SE +/- 3717554.45, N = 3SE +/- 940265.92, N = 3SE +/- 749718.47, N = 3SE +/- 147685.85, N = 3SE +/- 6141471.96, N = 3SE +/- 1289603.73, N = 3SE +/- 4432294.36, N = 3SE +/- 1611276.24, N = 3SE +/- 3610015.39, N = 3SE +/- 124677.36, N = 3SE +/- 584646.14, N = 3SE +/- 271682.00, N = 3SE +/- 958772.36, N = 3SE +/- 673003.22, N = 3SE +/- 3039034.93, N = 3SE +/- 752558.60, N = 3SE +/- 2261241.06, N = 3SE +/- 1678355.68, N = 3SE +/- 2292977.20, N = 3SE +/- 822739.33, N = 38837012507837700007994966677444100006465133335032533337571766676890033337325600006323966673844066673985666673497466673595766677093933337096500008493400008505866678245833338113733335772633336338600001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 4KXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 2.84, N = 15SE +/- 2.83, N = 15SE +/- 2.18, N = 15SE +/- 2.31, N = 15SE +/- 3.34, N = 15SE +/- 1.11, N = 15SE +/- 1.67, N = 15SE +/- 1.62, N = 15SE +/- 1.63, N = 15SE +/- 1.44, N = 15SE +/- 1.79, N = 15SE +/- 1.12, N = 7SE +/- 0.80, N = 15SE +/- 1.01, N = 8SE +/- 4.46, N = 15SE +/- 4.48, N = 15SE +/- 1.54, N = 15SE +/- 3.81, N = 15SE +/- 4.23, N = 15SE +/- 4.38, N = 15SE +/- 3.84, N = 15SE +/- 4.25, N = 15162.98174.83154.50162.47138.72133.19124.26150.00122.62144.70116.62124.60108.29114.54159.75163.41161.34162.48167.22172.88153.42160.791. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: SlowXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1020304050SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.18, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.23, N = 4SE +/- 0.04, N = 3SE +/- 0.12, N = 4SE +/- 0.04, N = 3SE +/- 0.21, N = 4SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 327.1116.0624.7115.7113.7910.9322.1413.8919.7513.3621.6315.2420.4613.7737.1329.0043.3929.1840.9628.0115.7717.04

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.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 32Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P9M18M27M36M45MSE +/- 419600.12, N = 5SE +/- 417042.80, N = 5SE +/- 1154.70, N = 3SE +/- 1527.53, N = 3SE +/- 577.35, N = 3SE +/- 1000.00, N = 3SE +/- 1452.97, N = 3SE +/- 3000.00, N = 3SE +/- 1201.85, N = 3SE +/- 1666.67, N = 3SE +/- 881.92, N = 3SE +/- 1527.53, N = 3SE +/- 881.92, N = 3SE +/- 0.00, N = 3SE +/- 2185.81, N = 3SE +/- 6984.11, N = 3SE +/- 19139.84, N = 3SE +/- 7023.77, N = 3SE +/- 13503.09, N = 3SE +/- 7753.14, N = 3SE +/- 881.92, N = 3SE +/- 2333.33, N = 3389506004024400036461000365060003199200031995000365213333653900036497667365376673519066735259000317056673172600029893333299073333604000036066000360330003606833329899333298996671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: GoogLeNetXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P80160240320400SE +/- 0.21, N = 3SE +/- 1.12, N = 3SE +/- 2.37, N = 3SE +/- 1.62, N = 3SE +/- 1.42, N = 3SE +/- 0.49, N = 3SE +/- 2.38, N = 3SE +/- 0.83, N = 3SE +/- 3.52, N = 5SE +/- 1.94, N = 3SE +/- 0.21, N = 3SE +/- 0.64, N = 3SE +/- 2.29, N = 3SE +/- 0.29, N = 3SE +/- 1.18, N = 3SE +/- 3.84, N = 4SE +/- 2.47, N = 3SE +/- 1.66, N = 3SE +/- 2.41, N = 3SE +/- 4.22, N = 3SE +/- 0.44, N = 3SE +/- 0.30, N = 3291.37319.76331.76380.14347.72282.50308.95335.58341.51333.73190.80232.51168.10214.14176.00314.67248.03330.26227.22354.59256.79276.93

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.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 57Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P600M1200M1800M2400M3000MSE +/- 32405829.38, N = 3SE +/- 11426868.92, N = 3SE +/- 28500058.48, N = 3SE +/- 7076957.92, N = 3SE +/- 25737283.81, N = 3SE +/- 4885465.97, N = 3SE +/- 27382304.50, N = 5SE +/- 5493127.02, N = 3SE +/- 28812627.44, N = 4SE +/- 2629533.12, N = 3SE +/- 7523814.04, N = 3SE +/- 1361779.88, N = 3SE +/- 4223873.95, N = 3SE +/- 1953913.45, N = 3SE +/- 17457503.96, N = 3SE +/- 8567444.84, N = 3SE +/- 8732315.46, N = 3SE +/- 8476044.68, N = 3SE +/- 10590929.03, N = 3SE +/- 3786085.64, N = 3SE +/- 7776531.65, N = 3SE +/- 4914717.13, N = 326200666672475500000265550000025392000001999266667146916666724913400002308966667249405000021387666672036733333209903333318523666671906266667237476666724449333332772800000286200000027302333332859833333202633333321868333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 512Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P300M600M900M1200M1500MSE +/- 4790383.89, N = 3SE +/- 2313246.88, N = 3SE +/- 6842108.19, N = 3SE +/- 1740596.96, N = 3SE +/- 8061118.62, N = 3SE +/- 912164.70, N = 3SE +/- 2123152.79, N = 3SE +/- 1345523.44, N = 3SE +/- 11560043.25, N = 5SE +/- 681281.47, N = 3SE +/- 866544.86, N = 3SE +/- 287923.60, N = 3SE +/- 1072572.19, N = 3SE +/- 305632.31, N = 3SE +/- 503322.30, N = 3SE +/- 814452.78, N = 3SE +/- 3901424.24, N = 3SE +/- 2412697.89, N = 3SE +/- 2338327.99, N = 3SE +/- 1047748.91, N = 3SE +/- 431058.13, N = 3SE +/- 390953.25, N = 3149383333311529333331126466667918566667981090000539226667120606666788220000010914600008001866677779400007963400007070533337196266671349800000136980000013947333331368733333119996666711581666677317566678099166671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 256 - Buffer Length: 256 - Filter Length: 512Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P500M1000M1500M2000M2500MSE +/- 14465476.14, N = 3SE +/- 240370.09, N = 3SE +/- 5630373.98, N = 3SE +/- 1722723.94, N = 3SE +/- 11053858.05, N = 3SE +/- 4629652.01, N = 3SE +/- 6728628.72, N = 3SE +/- 740202.67, N = 3SE +/- 4238841.56, N = 3SE +/- 569766.03, N = 3SE +/- 6411708.04, N = 3SE +/- 409267.64, N = 3SE +/- 2198737.01, N = 3SE +/- 320537.74, N = 3SE +/- 1650252.51, N = 3SE +/- 1300000.00, N = 3SE +/- 14870701.99, N = 3SE +/- 2676440.42, N = 3SE +/- 11753202.87, N = 3SE +/- 520683.31, N = 3SE +/- 170391.71, N = 3SE +/- 776881.23, N = 32027000000117876666716169666671020833333106103333354260666715189333339445500001444066667838530000143710000094027000013067333338483766671734800000174470000023943333331570400000212953333313621666677417400008210133331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P90K180K270K360K450KSE +/- 288.38, N = 3SE +/- 46.28, N = 3SE +/- 292.24, N = 3SE +/- 10.67, N = 3SE +/- 12.67, N = 3SE +/- 6.33, N = 3SE +/- 576.80, N = 3SE +/- 8.67, N = 3SE +/- 866.68, N = 3SE +/- 8.84, N = 3SE +/- 2096.39, N = 3SE +/- 87.49, N = 3SE +/- 801.76, N = 3SE +/- 32.84, N = 3SE +/- 4958.68, N = 4SE +/- 1.33, N = 3SE +/- 1315.55, N = 3SE +/- 154.46, N = 3SE +/- 710.46, N = 3SE +/- 20.00, N = 3SE +/- 0.00, N = 3SE +/- 22.28, N = 320677810578118360394551113219569391559597914413386667875402275211311363133190058412723220188376268192898320527166541934721022221. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

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.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 57Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1000M2000M3000M4000M5000MSE +/- 51282832.98, N = 4SE +/- 5260228.13, N = 3SE +/- 13655808.69, N = 3SE +/- 3116800.35, N = 3SE +/- 11686221.71, N = 3SE +/- 1549551.91, N = 3SE +/- 14772609.79, N = 3SE +/- 3119294.79, N = 3SE +/- 25888414.40, N = 3SE +/- 2030052.00, N = 3SE +/- 13354940.83, N = 3SE +/- 1909915.65, N = 3SE +/- 2709448.49, N = 3SE +/- 6887912.12, N = 3SE +/- 14606429.33, N = 3SE +/- 18691531.06, N = 3SE +/- 11892621.43, N = 3SE +/- 6778479.02, N = 3SE +/- 4098915.85, N = 3SE +/- 20120967.50, N = 3SE +/- 6398003.16, N = 3SE +/- 3527510.93, N = 344505250003503200000372366666731812666672689033333155466666738965000002972400000359640000026700333334068666667413676666737121333333736300000462656666747145000004759366667475533333342809333334256300000268706666729828000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 4 - Input: Bosphorus 4KXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P246810SE +/- 0.044, N = 3SE +/- 0.005, N = 3SE +/- 0.049, N = 3SE +/- 0.030, N = 3SE +/- 0.014, N = 3SE +/- 0.014, N = 3SE +/- 0.042, N = 3SE +/- 0.047, N = 3SE +/- 0.015, N = 3SE +/- 0.027, N = 3SE +/- 0.041, N = 3SE +/- 0.003, N = 3SE +/- 0.023, N = 3SE +/- 0.002, N = 3SE +/- 0.004, N = 3SE +/- 0.024, N = 3SE +/- 0.040, N = 3SE +/- 0.021, N = 3SE +/- 0.046, N = 3SE +/- 0.019, N = 3SE +/- 0.015, N = 3SE +/- 0.007, N = 37.3607.8766.6987.0615.4375.4386.0526.5725.9916.4904.0404.3503.7103.9727.1537.2838.5138.5958.3498.4226.7276.9391. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: Asian Dragon ObjXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50100150200250SE +/- 0.29, N = 4SE +/- 0.35, N = 3SE +/- 0.22, N = 4SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 3SE +/- 0.89, N = 15SE +/- 0.09, N = 3SE +/- 0.76, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.22, N = 3SE +/- 0.03, N = 3SE +/- 3.26, N = 12SE +/- 0.07, N = 4SE +/- 1.72, N = 15SE +/- 0.11, N = 4SE +/- 1.58, N = 15SE +/- 0.02, N = 4SE +/- 0.04, N = 3SE +/- 0.04, N = 3173.2297.19132.2879.3884.8047.64112.7271.83104.8464.96136.2283.73125.1176.53200.16133.93210.44133.48175.60113.2863.3464.33

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P90K180K270K360K450KSE +/- 649.08, N = 3SE +/- 79.75, N = 3SE +/- 348.45, N = 3SE +/- 11.39, N = 3SE +/- 14.62, N = 3SE +/- 6.33, N = 3SE +/- 169.28, N = 3SE +/- 8.67, N = 3SE +/- 116.69, N = 3SE +/- 17.33, N = 3SE +/- 482.17, N = 3SE +/- 14.67, N = 3SE +/- 353.58, N = 3SE +/- 174.44, N = 3SE +/- 2397.15, N = 3SE +/- 1.33, N = 3SE +/- 454.86, N = 3SE +/- 98.33, N = 3SE +/- 3455.48, N = 3SE +/- 38.33, N = 3SE +/- 20.54, N = 3SE +/- 13.72, N = 320549910583118358394513113223569391566587915213456267893404739211449363300190209415412220188374589192574316492166541934731021981. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

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.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 256 - Buffer Length: 256 - Filter Length: 57Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2000M4000M6000M8000M10000MSE +/- 61971606.40, N = 3SE +/- 3555434.03, N = 3SE +/- 2331189.49, N = 3SE +/- 5292237.50, N = 3SE +/- 3572580.77, N = 3SE +/- 2935605.18, N = 3SE +/- 28453880.81, N = 3SE +/- 1976810.00, N = 3SE +/- 17558125.69, N = 3SE +/- 4088330.28, N = 3SE +/- 30821331.00, N = 3SE +/- 6906036.01, N = 3SE +/- 28436830.58, N = 3SE +/- 8015887.00, N = 3SE +/- 8069352.17, N = 3SE +/- 12739745.33, N = 3SE +/- 50144890.07, N = 3SE +/- 8144527.82, N = 3SE +/- 60380193.59, N = 3SE +/- 2618735.91, N = 3SE +/- 4481443.22, N = 3SE +/- 1959875.28, N = 362843000003806833333536126666735008333333037100000161046666750360000003222533333482513333328489333337597566667471470000068009000004259433333631626666764013666678012700000581190000074118666675266466667279370000031169333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 256 - Buffer Length: 256 - Filter Length: 32Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2000M4000M6000M8000M10000MSE +/- 6711433.03, N = 3SE +/- 2516611.48, N = 3SE +/- 10496401.50, N = 3SE +/- 4021608.30, N = 3SE +/- 7369833.41, N = 3SE +/- 1266666.67, N = 3SE +/- 895048.11, N = 3SE +/- 1338323.99, N = 3SE +/- 2447674.63, N = 3SE +/- 1320353.49, N = 3SE +/- 20501246.57, N = 3SE +/- 2601495.30, N = 3SE +/- 32054242.22, N = 3SE +/- 821245.67, N = 3SE +/- 2936551.27, N = 3SE +/- 3768731.00, N = 3SE +/- 34610274.65, N = 3SE +/- 11426625.83, N = 3SE +/- 46730658.51, N = 3SE +/- 1419311.26, N = 3SE +/- 1928730.15, N = 3SE +/- 1501110.70, N = 360773000003167200000611656666732438000003572233333182356666753670333332769066667467096666723818000007546366667479913333368215333334301166667643900000065882000007935266667586246666775133333335271933333278960000030622000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 32Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P900M1800M2700M3600M4500MSE +/- 13813198.20, N = 3SE +/- 2236316.42, N = 3SE +/- 26633124.74, N = 3SE +/- 1858314.65, N = 3SE +/- 3557152.42, N = 3SE +/- 3178574.10, N = 3SE +/- 2403700.85, N = 3SE +/- 1342054.81, N = 3SE +/- 4598671.31, N = 3SE +/- 400000.00, N = 3SE +/- 517472.49, N = 3SE +/- 2023198.79, N = 3SE +/- 3194265.70, N = 3SE +/- 920748.49, N = 3SE +/- 384418.75, N = 3SE +/- 4056818.68, N = 3SE +/- 4250882.26, N = 3SE +/- 841295.30, N = 3SE +/- 3186603.90, N = 3SE +/- 2714365.08, N = 3SE +/- 2003607.86, N = 3SE +/- 3138116.92, N = 335207666673148233333355860000032006000003044700000179740000033370666672739633333325403333323519000004021033333410220000036079000003671966667379413333337915666674407900000435353333340633333334032733333276793333330470666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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.

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2004006008001000SE +/- 0.38, N = 3SE +/- 0.13, N = 3SE +/- 1.51, N = 3SE +/- 0.22, N = 3SE +/- 0.53, N = 3SE +/- 7.09, N = 3SE +/- 0.48, N = 3SE +/- 0.24, N = 3SE +/- 0.30, N = 3SE +/- 0.26, N = 3SE +/- 0.06, N = 3SE +/- 0.18, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.88, N = 3SE +/- 0.45, N = 3SE +/- 0.72, N = 3SE +/- 0.69, N = 3SE +/- 0.56, N = 3SE +/- 0.55, N = 3SE +/- 0.60, N = 3SE +/- 0.14, N = 3799.49799.72712.70711.38548.36555.60712.09712.01710.92711.10538.91540.24494.38494.05655.05655.80783.20783.42782.32783.78656.25655.241. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

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.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 32Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P500M1000M1500M2000M2500MSE +/- 17236814.61, N = 3SE +/- 27191440.65, N = 3SE +/- 3300505.01, N = 3SE +/- 3167192.94, N = 3SE +/- 9052685.30, N = 3SE +/- 3209534.41, N = 3SE +/- 3017909.50, N = 3SE +/- 692820.32, N = 3SE +/- 2206304.10, N = 3SE +/- 688799.28, N = 3SE +/- 4028647.42, N = 3SE +/- 808290.38, N = 3SE +/- 1003881.36, N = 3SE +/- 1379613.47, N = 3SE +/- 1462114.15, N = 3SE +/- 1550627.11, N = 3SE +/- 2510644.01, N = 3SE +/- 3000185.18, N = 3SE +/- 1567730.14, N = 3SE +/- 2810891.52, N = 3SE +/- 3256958.36, N = 3SE +/- 3090487.20, N = 321546333331927766667202500000018905666671731166667156176666720904666671679700000196323333316402666671989300000205730000017805333331841400000189496666718987666672279300000228746666722820666672282766667177686666718713333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Very FastXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.16, N = 5SE +/- 0.08, N = 3SE +/- 0.50, N = 5SE +/- 0.01, N = 4SE +/- 0.18, N = 4SE +/- 0.06, N = 3SE +/- 0.38, N = 9SE +/- 0.02, N = 4SE +/- 0.39, N = 4SE +/- 0.12, N = 4SE +/- 0.43, N = 15SE +/- 0.06, N = 4SE +/- 0.34, N = 15SE +/- 0.01, N = 4SE +/- 0.43, N = 15SE +/- 0.18, N = 5SE +/- 0.12, N = 5SE +/- 0.13, N = 5SE +/- 0.55, N = 5SE +/- 0.13, N = 5SE +/- 0.06, N = 4SE +/- 0.08, N = 453.6635.1054.8849.4240.6930.1746.2540.8244.7539.5628.4344.4427.3340.7346.0054.4655.0562.7053.4462.0244.9250.04

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Junkshop - Compute: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.24, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.19, N = 3SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.14, N = 3SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 4SE +/- 0.08, N = 3SE +/- 0.08, N = 4SE +/- 0.13, N = 3SE +/- 0.02, N = 4SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 318.3132.0821.1535.9633.2061.2223.4539.1724.9942.8516.8926.4918.7029.0611.2117.7011.4718.6512.8721.0238.8635.82

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Fishy Cat - Compute: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1326395265SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.11, N = 3SE +/- 0.27, N = 3SE +/- 0.15, N = 3SE +/- 0.18, N = 3SE +/- 0.04, N = 3SE +/- 0.13, N = 3SE +/- 0.17, N = 3SE +/- 0.01, N = 4SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.37, N = 3SE +/- 0.04, N = 5SE +/- 0.13, N = 3SE +/- 0.02, N = 5SE +/- 0.07, N = 3SE +/- 0.05, N = 4SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.23, N = 317.2931.9119.1334.8330.7559.1521.4538.4823.1742.7815.1626.5216.5229.559.7116.5810.1618.3611.2520.1538.0434.35

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Disney MaterialXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780EXeon 6766EEPYC 9754EPYC 9684XEPYC 9654EPYC 8534PNEPYC 8534P2040608010043.7048.4348.5653.4958.3879.8349.1456.4051.2558.3852.3258.3145.0840.9243.4561.3656.42

Scene: Disney Material

EPYC 9684X 2P: The test run did not produce a result.

EPYC 9654 2P: The test run did not produce a result.

EPYC 9754 2P: The test run did not produce a result.

Xeon 6766E 2P: The test run did not produce a result.

Xeon 6780E 2P: The test run did not produce a result.

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: AlexNetXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2004006008001000SE +/- 2.22, N = 5SE +/- 6.22, N = 5SE +/- 7.98, N = 5SE +/- 0.51, N = 5SE +/- 3.70, N = 5SE +/- 0.27, N = 4SE +/- 7.35, N = 5SE +/- 2.31, N = 5SE +/- 6.39, N = 13SE +/- 1.95, N = 5SE +/- 5.91, N = 15SE +/- 0.97, N = 3SE +/- 5.44, N = 15SE +/- 0.70, N = 3SE +/- 7.43, N = 15SE +/- 2.99, N = 5SE +/- 7.66, N = 5SE +/- 1.37, N = 5SE +/- 8.87, N = 4SE +/- 1.31, N = 5SE +/- 0.16, N = 4SE +/- 0.45, N = 4858.84819.89996.47969.62904.36690.09940.40852.36932.59841.68520.01482.06464.80443.74658.97924.80738.31868.54727.77942.19617.03673.42

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: MediumXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.27, N = 4SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.23, N = 3SE +/- 0.01, N = 3SE +/- 0.21, N = 3SE +/- 0.00, N = 3SE +/- 0.52, N = 15SE +/- 0.04, N = 4SE +/- 0.40, N = 5SE +/- 0.01, N = 4SE +/- 0.56, N = 5SE +/- 0.07, N = 4SE +/- 0.03, N = 3SE +/- 0.05, N = 341.5824.8032.0420.9720.3315.5129.6218.8426.6918.1331.7021.3929.1519.3454.5739.0863.4840.5061.1139.5823.0224.311. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.018471, N = 4SE +/- 0.002400, N = 3SE +/- 0.016400, N = 4SE +/- 0.002987, N = 3SE +/- 0.052953, N = 3SE +/- 0.034794, N = 3SE +/- 0.003253, N = 4SE +/- 0.013353, N = 3SE +/- 0.003237, N = 4SE +/- 0.001847, N = 3SE +/- 0.037087, N = 15SE +/- 0.074320, N = 15SE +/- 0.041359, N = 15SE +/- 0.080453, N = 15SE +/- 0.040626, N = 15SE +/- 0.052975, N = 7SE +/- 0.029749, N = 4SE +/- 0.024479, N = 49.96452616.96910011.28438016.03080015.24719025.78945012.10690019.88246013.21125021.0211202.9404565.1476583.0811085.4606603.4083375.91645812.15132011.3696401. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: SlowXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1428425670SE +/- 0.04, N = 4SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.18, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.47, N = 8SE +/- 0.05, N = 4SE +/- 0.26, N = 5SE +/- 0.07, N = 4SE +/- 0.50, N = 5SE +/- 0.01, N = 4SE +/- 0.05, N = 3SE +/- 0.04, N = 340.9224.2331.2420.2319.7115.0928.9618.2426.0017.5231.8520.8729.1518.8654.8838.4762.8240.0061.0738.9022.4623.711. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -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.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput ThreadXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3129.5129.5129.7129.7112.3112.3129.7129.7129.7129.752.652.549.849.8112.1112.1112.2112.2112.2112.2112.1112.11. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Very FastXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20406080100SE +/- 0.36, N = 5SE +/- 0.04, N = 4SE +/- 0.45, N = 5SE +/- 0.08, N = 4SE +/- 0.18, N = 4SE +/- 0.06, N = 3SE +/- 0.29, N = 4SE +/- 0.08, N = 4SE +/- 0.33, N = 4SE +/- 0.06, N = 4SE +/- 0.50, N = 15SE +/- 0.00, N = 4SE +/- 0.36, N = 15SE +/- 0.04, N = 4SE +/- 0.63, N = 15SE +/- 0.21, N = 5SE +/- 0.28, N = 6SE +/- 0.32, N = 6SE +/- 0.71, N = 6SE +/- 0.62, N = 6SE +/- 0.07, N = 4SE +/- 0.07, N = 565.3838.0267.6450.0344.1131.2552.2341.3451.9140.2338.9246.1836.4041.8659.0168.5674.0378.1572.6876.8351.5154.601. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution TimeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2040608010023.5633.0728.5642.2240.2974.9627.5841.7528.9945.9524.5733.8122.7128.3325.0132.5959.2758.171. (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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh TimeXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P91827364529.3331.7035.6031.3836.8639.1228.8432.0032.4831.7829.4730.3826.8225.9327.7428.0336.8233.671. (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

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 8 - Input: Bosphorus 4KXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20406080100SE +/- 0.79, N = 5SE +/- 0.22, N = 5SE +/- 0.46, N = 4SE +/- 0.20, N = 5SE +/- 0.52, N = 7SE +/- 0.09, N = 4SE +/- 0.44, N = 15SE +/- 0.29, N = 4SE +/- 0.58, N = 5SE +/- 0.26, N = 4SE +/- 0.34, N = 3SE +/- 0.02, N = 4SE +/- 0.35, N = 15SE +/- 0.03, N = 4SE +/- 0.31, N = 4SE +/- 0.38, N = 5SE +/- 0.63, N = 5SE +/- 0.42, N = 5SE +/- 0.40, N = 5SE +/- 0.56, N = 5SE +/- 0.67, N = 4SE +/- 0.11, N = 470.4379.4465.2270.9159.1055.2455.9865.3654.8262.7547.2550.8242.8646.4877.6078.7487.8286.3086.3487.9964.7968.191. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.2Run: RTLightmap.hdr.4096x4096 - Device: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.541.081.622.162.7SE +/- 0.01, N = 4SE +/- 0.00, N = 3SE +/- 0.01, N = 4SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 4SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.401.582.181.291.480.861.741.141.721.081.681.071.510.981.711.501.641.281.671.310.800.86

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: BMW27 - Compute: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1122334455SE +/- 0.03, N = 4SE +/- 0.04, N = 3SE +/- 0.08, N = 4SE +/- 0.15, N = 3SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 4SE +/- 0.24, N = 3SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.09, N = 5SE +/- 0.05, N = 3SE +/- 0.06, N = 5SE +/- 0.17, N = 8SE +/- 0.02, N = 6SE +/- 0.01, N = 4SE +/- 0.03, N = 6SE +/- 0.04, N = 4SE +/- 0.02, N = 5SE +/- 0.07, N = 4SE +/- 0.05, N = 3SE +/- 0.04, N = 312.7623.7213.7425.8923.6746.3215.2928.4916.5531.669.7117.6110.5619.857.1712.867.7514.298.6115.9630.0727.12

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.

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50100150200250SE +/- 1.02, N = 8SE +/- 0.20, N = 6SE +/- 0.81, N = 7SE +/- 0.61, N = 5SE +/- 1.19, N = 15SE +/- 0.38, N = 9SE +/- 1.85, N = 12SE +/- 0.84, N = 15SE +/- 1.28, N = 12SE +/- 0.12, N = 4SE +/- 0.46, N = 15SE +/- 0.01, N = 3SE +/- 0.05, N = 5SE +/- 0.02, N = 3SE +/- 3.98, N = 15SE +/- 0.70, N = 9SE +/- 6.15, N = 12SE +/- 3.27, N = 13SE +/- 2.97, N = 13SE +/- 1.18, N = 7SE +/- 0.03, N = 6SE +/- 0.04, N = 6199.73106.40133.6382.8585.2947.26114.9070.84108.4063.6284.3543.5576.6039.17214.62229.91236.61151.95217.85136.9598.21107.351. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P13002600390052006500SE +/- 25.44, N = 8SE +/- 5.07, N = 6SE +/- 20.15, N = 7SE +/- 15.23, N = 5SE +/- 29.84, N = 15SE +/- 9.38, N = 9SE +/- 46.19, N = 12SE +/- 21.10, N = 15SE +/- 32.01, N = 12SE +/- 2.91, N = 4SE +/- 11.58, N = 15SE +/- 0.29, N = 3SE +/- 1.36, N = 5SE +/- 0.46, N = 3SE +/- 99.42, N = 15SE +/- 17.54, N = 9SE +/- 153.85, N = 12SE +/- 81.83, N = 13SE +/- 74.36, N = 13SE +/- 29.52, N = 7SE +/- 0.82, N = 6SE +/- 0.97, N = 64993.342659.883340.692071.192132.241181.492872.561771.072709.931590.502108.681088.831915.04979.285365.425747.825915.253798.705446.163423.632455.162683.831. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.6Video Input: Bosphorus 4KXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P918273645SE +/- 0.09, N = 4SE +/- 0.06, N = 4SE +/- 0.20, N = 4SE +/- 0.11, N = 4SE +/- 0.05, N = 3SE +/- 0.13, N = 3SE +/- 0.10, N = 3SE +/- 0.39, N = 3SE +/- 0.12, N = 3SE +/- 0.08, N = 3SE +/- 0.17, N = 3SE +/- 0.13, N = 3SE +/- 0.22, N = 6SE +/- 0.11, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 339.3138.9337.0437.8630.8029.7627.9031.4027.2629.0519.5521.3022.0720.2228.1128.1631.9532.1228.8131.7227.7428.131. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Timed FFmpeg Compilation

This test times how long it takes to build the FFmpeg multimedia library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 7.0Time To CompileXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P816243240SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.30, N = 3SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 317.3921.3619.5124.0625.9434.2920.0124.8420.7926.4623.3225.2425.7127.8521.2223.2217.8419.9918.0920.4928.4227.54

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.013568, N = 6SE +/- 0.022698, N = 7SE +/- 0.019069, N = 15SE +/- 0.022393, N = 7SE +/- 0.037316, N = 9SE +/- 0.010812, N = 4SE +/- 0.008354, N = 7SE +/- 0.023904, N = 7SE +/- 0.010204, N = 7SE +/- 0.025580, N = 6SE +/- 0.022246, N = 15SE +/- 0.115702, N = 15SE +/- 0.055800, N = 15SE +/- 0.115724, N = 15SE +/- 0.063684, N = 15SE +/- 0.115723, N = 15SE +/- 0.045713, N = 13SE +/- 0.057580, N = 62.1594594.1827072.5298074.9558654.70255711.8521302.5273635.1141612.8810995.9309572.0843843.6543032.0768353.3604532.3129773.7548356.8527126.5417681. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

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.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 1 - Model: AlexNetXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1326395265SE +/- 0.45, N = 15SE +/- 0.66, N = 15SE +/- 0.44, N = 15SE +/- 0.08, N = 8SE +/- 0.33, N = 15SE +/- 0.05, N = 7SE +/- 0.52, N = 15SE +/- 1.22, N = 15SE +/- 0.57, N = 15SE +/- 0.76, N = 15SE +/- 0.19, N = 5SE +/- 0.08, N = 7SE +/- 0.12, N = 15SE +/- 0.07, N = 6SE +/- 0.15, N = 9SE +/- 0.48, N = 15SE +/- 0.10, N = 6SE +/- 0.07, N = 7SE +/- 0.17, N = 15SE +/- 0.18, N = 7SE +/- 0.04, N = 7SE +/- 0.05, N = 730.7635.5037.5756.5337.2642.0533.2143.6737.7648.2818.3431.9916.2429.5618.5935.0819.9226.2319.8832.5634.3534.51

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P80K160K240K320K400KSE +/- 1528.09, N = 5SE +/- 31.76, N = 4SE +/- 560.20, N = 4SE +/- 150.79, N = 3SE +/- 33.85, N = 3SE +/- 11.99, N = 3SE +/- 420.81, N = 4SE +/- 9.58, N = 3SE +/- 679.14, N = 4SE +/- 67.02, N = 3SE +/- 2064.71, N = 6SE +/- 126.50, N = 4SE +/- 1666.53, N = 6SE +/- 1178.78, N = 6SE +/- 1929.89, N = 5SE +/- 72.03, N = 4SE +/- 14.31, N = 3SE +/- 23.20, N = 3228184.68109416.42147311.1471245.2690420.2537567.61141954.1569776.41135939.2558815.36214017.18127519.37390520.93225311.82256584.67125191.1152684.9352745.641. (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

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.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P2004006008001000SE +/- 0.48, N = 4SE +/- 0.41, N = 4SE +/- 1.15, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 2.85, N = 3SE +/- 0.67, N = 3SE +/- 0.58, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 1.15, N = 3SE +/- 1.33, N = 3SE +/- 0.58, N = 3SE +/- 1.20, N = 3SE +/- 2.08, N = 3SE +/- 1.86, N = 3SE +/- 2.19, N = 3SE +/- 2.40, N = 3SE +/- 1.45, N = 36926917677659719737637627637631005100711191117972974818818822818974976

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P130K260K390K520K650KSE +/- 726.00, N = 5SE +/- 213.44, N = 5SE +/- 398.40, N = 5SE +/- 130.20, N = 4SE +/- 155.95, N = 4SE +/- 123.65, N = 3SE +/- 411.20, N = 5SE +/- 37.04, N = 4SE +/- 441.66, N = 5SE +/- 71.37, N = 3SE +/- 5800.63, N = 15SE +/- 2566.89, N = 15SE +/- 1879.23, N = 6SE +/- 1849.65, N = 6SE +/- 4162.49, N = 6SE +/- 789.76, N = 5SE +/- 274.93, N = 3SE +/- 314.42, N = 4455312.09232012.77261127.09135125.15185954.5498557.16257384.55131606.86246822.52126171.52588329.32279397.45551955.55362511.43472931.04269047.28130839.12133284.461. (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

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.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ThoroughXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P306090120150SE +/- 0.55, N = 5SE +/- 0.05, N = 4SE +/- 0.19, N = 5SE +/- 0.01, N = 4SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 5SE +/- 0.03, N = 4SE +/- 0.05, N = 4SE +/- 0.01, N = 3SE +/- 0.85, N = 4SE +/- 0.01, N = 3SE +/- 0.42, N = 4SE +/- 0.00, N = 3SE +/- 1.20, N = 5SE +/- 0.01, N = 5SE +/- 1.46, N = 12SE +/- 0.06, N = 5SE +/- 0.73, N = 15SE +/- 0.01, N = 4SE +/- 0.02, N = 3SE +/- 0.02, N = 497.9454.4586.9851.3030.0315.5083.3442.9872.3337.0179.2945.2171.0240.69121.3086.01116.1273.38105.4165.3037.9641.021. (CXX) g++ options: -O3 -flto -pthread

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3K6K9K12K15KSE +/- 155.00, N = 15SE +/- 8.92, N = 4SE +/- 23.92, N = 4SE +/- 6.56, N = 3SE +/- 8.35, N = 4SE +/- 11.05, N = 3SE +/- 4.96, N = 4SE +/- 18.08, N = 4SE +/- 12.99, N = 4SE +/- 15.55, N = 4SE +/- 85.04, N = 8SE +/- 35.67, N = 5SE +/- 119.22, N = 6SE +/- 29.95, N = 5SE +/- 76.70, N = 10SE +/- 17.53, N = 5SE +/- 14.39, N = 4SE +/- 9.85, N = 47437.814059.703960.282865.723345.452266.323819.833076.663816.703050.169771.415702.1312296.276346.1510139.695229.412718.942746.201. (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

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 24.0Time To CompileXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.05, N = 4SE +/- 0.07, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.17, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 4SE +/- 0.02, N = 3SE +/- 0.02, N = 4SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 316.1817.2618.2919.2321.4423.6618.1419.6518.4720.1823.4623.4125.8825.8919.5619.9116.4016.6716.4116.9621.5321.04

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.03329814, N = 5SE +/- 0.02923582, N = 4SE +/- 0.04312801, N = 5SE +/- 0.01004100, N = 4SE +/- 0.02691998, N = 4SE +/- 0.01216869, N = 3SE +/- 0.05695382, N = 15SE +/- 0.01708632, N = 4SE +/- 0.01741826, N = 5SE +/- 0.04445674, N = 4SE +/- 0.01633300, N = 4SE +/- 0.03951909, N = 5SE +/- 0.00891960, N = 5SE +/- 0.05847189, N = 6SE +/- 0.04312156, N = 5SE +/- 0.06577912, N = 5SE +/- 0.10237421, N = 3SE +/- 0.01040414, N = 35.3912596710.196706606.3916981712.9293413011.2498016023.970828406.6638751012.799758207.2934911713.969237604.510618938.872259143.753199016.881212794.260785968.2538351121.1780307020.937913901. (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

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.3Pi Digits To Calculate: 1BXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.010, N = 6SE +/- 0.030, N = 5SE +/- 0.007, N = 6SE +/- 0.005, N = 5SE +/- 0.010, N = 5SE +/- 0.011, N = 4SE +/- 0.142, N = 12SE +/- 0.026, N = 5SE +/- 0.122, N = 15SE +/- 0.045, N = 5SE +/- 0.002, N = 4SE +/- 0.020, N = 4SE +/- 0.007, N = 4SE +/- 0.018, N = 4SE +/- 0.049, N = 4SE +/- 0.007, N = 4SE +/- 0.012, N = 4SE +/- 0.008, N = 4SE +/- 0.012, N = 4SE +/- 0.008, N = 5SE +/- 0.001, N = 4SE +/- 0.005, N = 45.1106.7015.0087.0477.74410.7126.2667.9596.3348.3007.46810.3867.96711.1987.9156.9288.9927.4718.1966.74310.42710.205

Timed ImageMagick Compilation

This test times how long it takes to build ImageMagick. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.9.0Time To CompileXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P510152025SE +/- 0.03, N = 4SE +/- 0.10, N = 4SE +/- 0.13, N = 4SE +/- 0.10, N = 4SE +/- 0.13, N = 4SE +/- 0.03, N = 3SE +/- 0.07, N = 4SE +/- 0.06, N = 4SE +/- 0.11, N = 4SE +/- 0.15, N = 4SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 4SE +/- 0.08, N = 4SE +/- 0.04, N = 4SE +/- 0.06, N = 4SE +/- 0.04, N = 4SE +/- 0.09, N = 4SE +/- 0.04, N = 3SE +/- 0.04, N = 311.4912.6013.2614.7216.3318.8414.0915.3814.4016.0818.8519.2020.5920.9115.8116.3911.7212.2012.7513.4718.1817.92

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.2Time To SolveXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P612182430SE +/- 0.028, N = 6SE +/- 0.030, N = 4SE +/- 0.021, N = 6SE +/- 0.025, N = 4SE +/- 0.066, N = 4SE +/- 0.023, N = 3SE +/- 0.028, N = 5SE +/- 0.035, N = 3SE +/- 0.021, N = 5SE +/- 0.031, N = 3SE +/- 0.034, N = 7SE +/- 0.032, N = 5SE +/- 0.035, N = 7SE +/- 0.039, N = 5SE +/- 0.031, N = 8SE +/- 0.021, N = 6SE +/- 0.064, N = 15SE +/- 0.025, N = 6SE +/- 0.093, N = 15SE +/- 0.019, N = 6SE +/- 0.029, N = 4SE +/- 0.019, N = 47.40214.2577.88115.21112.39024.1759.28518.10210.84021.0305.51210.3016.04511.4793.9816.3504.1336.9504.6097.73814.49313.2501. (CXX) g++ options: -fopenmp -O2 -march=native

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.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.2Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1.14532.29063.43594.58125.7265SE +/- 0.03, N = 6SE +/- 0.00, N = 5SE +/- 0.01, N = 6SE +/- 0.00, N = 4SE +/- 0.00, N = 5SE +/- 0.00, N = 3SE +/- 0.02, N = 5SE +/- 0.00, N = 4SE +/- 0.02, N = 5SE +/- 0.00, N = 4SE +/- 0.00, N = 5SE +/- 0.00, N = 4SE +/- 0.01, N = 4SE +/- 0.00, N = 4SE +/- 0.01, N = 5SE +/- 0.03, N = 6SE +/- 0.00, N = 5SE +/- 0.02, N = 4SE +/- 0.01, N = 5SE +/- 0.01, N = 4SE +/- 0.00, N = 3SE +/- 0.00, N = 35.093.324.482.713.101.813.662.393.492.273.312.112.971.933.573.163.472.663.492.731.691.80

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.2Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-OnlyXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1.13182.26363.39544.52725.659SE +/- 0.03, N = 6SE +/- 0.00, N = 5SE +/- 0.03, N = 6SE +/- 0.00, N = 4SE +/- 0.01, N = 5SE +/- 0.00, N = 3SE +/- 0.04, N = 5SE +/- 0.00, N = 4SE +/- 0.01, N = 5SE +/- 0.00, N = 4SE +/- 0.03, N = 5SE +/- 0.00, N = 4SE +/- 0.00, N = 4SE +/- 0.00, N = 4SE +/- 0.01, N = 5SE +/- 0.02, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 4SE +/- 0.01, N = 5SE +/- 0.01, N = 4SE +/- 0.00, N = 3SE +/- 0.00, N = 35.033.314.412.713.081.813.652.383.512.263.312.112.981.933.593.123.452.643.482.731.691.79

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20K40K60K80K100KSE +/- 271.54, N = 6SE +/- 328.18, N = 8SE +/- 178.23, N = 8SE +/- 243.41, N = 12SE +/- 28.62, N = 7SE +/- 34.34, N = 6SE +/- 560.41, N = 15SE +/- 260.16, N = 15SE +/- 545.92, N = 15SE +/- 313.88, N = 7SE +/- 545.82, N = 5SE +/- 480.71, N = 15SE +/- 661.05, N = 7SE +/- 386.70, N = 9SE +/- 779.47, N = 7SE +/- 393.67, N = 15SE +/- 207.19, N = 6SE +/- 126.79, N = 698764.9363760.4558128.2433811.6338906.1322655.3266492.4435234.0665396.2235116.5563265.0143799.7888487.9162071.1084477.7253182.7721684.8022229.941. (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

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterGNU Octave Benchmark 8.4.0Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.019, N = 6SE +/- 0.020, N = 6SE +/- 0.019, N = 6SE +/- 0.015, N = 6SE +/- 0.052, N = 5SE +/- 0.017, N = 5SE +/- 0.082, N = 5SE +/- 0.024, N = 6SE +/- 0.096, N = 5SE +/- 0.015, N = 6SE +/- 0.108, N = 7SE +/- 0.051, N = 5SE +/- 0.023, N = 5SE +/- 0.035, N = 5SE +/- 0.025, N = 5SE +/- 0.014, N = 5SE +/- 0.007, N = 6SE +/- 0.018, N = 6SE +/- 0.004, N = 6SE +/- 0.009, N = 6SE +/- 0.007, N = 5SE +/- 0.011, N = 56.8566.4588.0367.5329.6949.10412.5787.74812.6177.62211.61910.24312.24510.9968.8488.8016.9556.8577.6457.5288.9488.929

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.3Pi Digits To Calculate: 500MXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1.16482.32963.49444.65925.824SE +/- 0.004, N = 7SE +/- 0.008, N = 7SE +/- 0.002, N = 8SE +/- 0.003, N = 7SE +/- 0.006, N = 7SE +/- 0.003, N = 6SE +/- 0.074, N = 15SE +/- 0.023, N = 15SE +/- 0.064, N = 15SE +/- 0.010, N = 7SE +/- 0.004, N = 6SE +/- 0.007, N = 6SE +/- 0.004, N = 6SE +/- 0.005, N = 6SE +/- 0.015, N = 5SE +/- 0.005, N = 6SE +/- 0.007, N = 5SE +/- 0.003, N = 6SE +/- 0.003, N = 5SE +/- 0.003, N = 6SE +/- 0.008, N = 6SE +/- 0.006, N = 62.6593.3912.3643.2673.7244.8742.9613.7043.1103.7813.4114.7703.6295.1774.1003.9684.5104.3124.3293.8505.1655.077

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: Asian DragonXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50100150200250SE +/- 0.26, N = 8SE +/- 0.13, N = 6SE +/- 0.13, N = 7SE +/- 0.05, N = 5SE +/- 0.29, N = 6SE +/- 0.04, N = 4SE +/- 1.20, N = 15SE +/- 0.11, N = 5SE +/- 0.63, N = 6SE +/- 0.04, N = 5SE +/- 0.16, N = 7SE +/- 0.05, N = 5SE +/- 0.10, N = 6SE +/- 0.04, N = 5SE +/- 1.88, N = 7SE +/- 0.12, N = 7SE +/- 2.04, N = 8SE +/- 0.07, N = 7SE +/- 1.54, N = 8SE +/- 0.03, N = 6SE +/- 0.05, N = 5SE +/- 0.03, N = 5199.61112.66152.7591.9999.2455.42129.7583.49125.7575.47151.2592.88138.8584.71208.31156.39231.43155.12185.71132.4973.9175.21

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.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Highest CompressionXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.81451.6292.44353.2584.0725SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 63.463.473.073.082.812.823.093.103.093.102.472.492.232.233.033.043.623.613.623.623.033.031. (CC) gcc options: -fvisibility=hidden -O2 -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.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: CrownXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50100150200250SE +/- 0.07, N = 7SE +/- 0.05, N = 5SE +/- 0.12, N = 6SE +/- 0.07, N = 5SE +/- 0.08, N = 5SE +/- 0.07, N = 3SE +/- 0.48, N = 6SE +/- 0.02, N = 4SE +/- 0.14, N = 6SE +/- 0.05, N = 4SE +/- 0.14, N = 6SE +/- 0.02, N = 5SE +/- 0.11, N = 6SE +/- 0.05, N = 5SE +/- 1.04, N = 7SE +/- 0.64, N = 6SE +/- 1.23, N = 8SE +/- 0.20, N = 7SE +/- 0.43, N = 7SE +/- 0.08, N = 6SE +/- 0.07, N = 4SE +/- 0.04, N = 4153.1682.97124.7170.0582.9243.67116.6165.33107.6358.42133.4281.83121.4373.96190.08126.15214.02130.79176.21110.5358.2862.19

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.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: MediumXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P120240360480600SE +/- 1.56, N = 7SE +/- 1.17, N = 6SE +/- 2.79, N = 6SE +/- 0.16, N = 6SE +/- 2.26, N = 5SE +/- 0.32, N = 4SE +/- 0.77, N = 6SE +/- 0.08, N = 6SE +/- 0.52, N = 6SE +/- 0.03, N = 6SE +/- 2.01, N = 5SE +/- 0.13, N = 6SE +/- 1.77, N = 5SE +/- 0.18, N = 5SE +/- 0.52, N = 6SE +/- 0.39, N = 7SE +/- 3.11, N = 7SE +/- 0.63, N = 7SE +/- 3.54, N = 6SE +/- 0.20, N = 7SE +/- 0.07, N = 6SE +/- 0.07, N = 6550.02327.78484.55333.99216.97126.44483.10284.33441.65247.30363.38314.62338.15284.46462.24527.80508.09500.77491.53449.11273.67294.621. (CXX) g++ options: -O3 -flto -pthread

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 12 - Input: Bosphorus 4KXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P4080120160200SE +/- 0.42, N = 7SE +/- 0.20, N = 7SE +/- 0.25, N = 7SE +/- 0.41, N = 7SE +/- 1.28, N = 6SE +/- 0.48, N = 6SE +/- 0.97, N = 6SE +/- 0.40, N = 7SE +/- 0.83, N = 5SE +/- 0.61, N = 7SE +/- 1.07, N = 6SE +/- 0.15, N = 6SE +/- 0.88, N = 5SE +/- 0.08, N = 6SE +/- 1.41, N = 6SE +/- 1.32, N = 6SE +/- 0.88, N = 6SE +/- 0.43, N = 7SE +/- 1.88, N = 15SE +/- 0.80, N = 7SE +/- 0.78, N = 6SE +/- 0.71, N = 6165.50177.86156.52164.25140.76134.07127.55151.72122.09146.10117.68126.59108.50115.35168.78169.23163.80165.75171.22178.52157.36165.091. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 6, LosslessXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P3691215SE +/- 0.031, N = 7SE +/- 0.025, N = 7SE +/- 0.022, N = 7SE +/- 0.013, N = 7SE +/- 0.027, N = 6SE +/- 0.016, N = 6SE +/- 0.035, N = 6SE +/- 0.033, N = 6SE +/- 0.027, N = 6SE +/- 0.045, N = 6SE +/- 0.038, N = 5SE +/- 0.010, N = 6SE +/- 0.054, N = 5SE +/- 0.012, N = 5SE +/- 0.025, N = 6SE +/- 0.007, N = 6SE +/- 0.015, N = 7SE +/- 0.014, N = 7SE +/- 0.020, N = 7SE +/- 0.010, N = 7SE +/- 0.011, N = 6SE +/- 0.009, N = 65.6135.7846.1726.2787.3888.2017.0386.6057.1216.8369.3518.3599.9609.1886.9866.8535.9505.7575.8465.7997.3017.1821. (CXX) g++ options: -O3 -fPIC -lm

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P20406080100SE +/- 0.11, N = 8SE +/- 0.08, N = 6SE +/- 0.28, N = 7SE +/- 0.09, N = 5SE +/- 0.09, N = 5SE +/- 0.01, N = 3SE +/- 0.04, N = 7SE +/- 0.09, N = 5SE +/- 0.06, N = 7SE +/- 0.04, N = 4SE +/- 0.04, N = 6SE +/- 0.01, N = 4SE +/- 0.03, N = 6SE +/- 0.01, N = 4SE +/- 0.19, N = 10SE +/- 0.17, N = 7SE +/- 0.39, N = 9SE +/- 0.40, N = 7SE +/- 0.10, N = 9SE +/- 0.22, N = 7SE +/- 0.10, N = 5SE +/- 0.16, N = 558.2529.1445.9723.5424.9713.4944.6722.5439.3519.7735.2017.8531.4815.9484.6944.9778.5045.2568.8739.0921.0021.821. (CC) gcc options: -O3 -march=native -fopenmp

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 10, LosslessXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P246810SE +/- 0.015, N = 8SE +/- 0.008, N = 8SE +/- 0.014, N = 7SE +/- 0.006, N = 7SE +/- 0.011, N = 7SE +/- 0.008, N = 7SE +/- 0.028, N = 7SE +/- 0.015, N = 7SE +/- 0.019, N = 7SE +/- 0.005, N = 7SE +/- 0.016, N = 6SE +/- 0.003, N = 6SE +/- 0.021, N = 6SE +/- 0.009, N = 6SE +/- 0.006, N = 7SE +/- 0.004, N = 7SE +/- 0.003, N = 7SE +/- 0.002, N = 8SE +/- 0.006, N = 8SE +/- 0.006, N = 8SE +/- 0.010, N = 7SE +/- 0.006, N = 74.5794.4565.0744.9235.9195.9065.3114.8675.3174.9167.6036.8498.0947.5075.7405.6624.9434.7704.8294.7925.7345.7351. (CXX) g++ options: -O3 -fPIC -lm

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P50K100K150K200K250KSE +/- 1422.07, N = 7SE +/- 406.29, N = 9SE +/- 1262.04, N = 9SE +/- 152.92, N = 9SE +/- 233.83, N = 9SE +/- 53.56, N = 9SE +/- 1298.67, N = 9SE +/- 382.59, N = 10SE +/- 1415.96, N = 9SE +/- 436.97, N = 10SE +/- 1776.37, N = 11SE +/- 884.47, N = 9SE +/- 7207.31, N = 15SE +/- 1372.89, N = 15SE +/- 2707.72, N = 14SE +/- 1153.71, N = 15SE +/- 301.54, N = 8SE +/- 167.65, N = 8220893.78107321.89163333.2183448.70110742.1557317.66171436.4786407.46172614.1085910.95245447.72125567.07244738.81134994.14194942.32116725.0462449.5563930.641. (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

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.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput ThreadXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P160320480640800SE +/- 1.89, N = 8SE +/- 1.34, N = 8SE +/- 2.57, N = 8SE +/- 1.17, N = 8SE +/- 2.08, N = 7SE +/- 3.57, N = 7SE +/- 5.02, N = 8SE +/- 1.27, N = 8SE +/- 3.86, N = 8SE +/- 1.64, N = 8SE +/- 0.66, N = 5SE +/- 0.24, N = 5SE +/- 0.32, N = 5SE +/- 0.32, N = 5SE +/- 4.48, N = 7SE +/- 0.80, N = 7SE +/- 5.52, N = 8SE +/- 3.13, N = 8SE +/- 3.72, N = 8SE +/- 1.71, N = 8SE +/- 1.24, N = 7SE +/- 3.14, N = 7755.5742.9675.5680.2490.3484.8663.8705.5651.9702.6323.7333.2293.1299.4576.8596.2664.7680.7673.4684.7563.6557.91. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

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.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: Rhodopsin ProteinXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468EPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1224364860SE +/- 0.97, N = 15SE +/- 0.29, N = 15SE +/- 0.33, N = 8SE +/- 0.36, N = 15SE +/- 0.53, N = 15SE +/- 0.28, N = 15SE +/- 0.09, N = 8SE +/- 0.04, N = 11SE +/- 0.11, N = 9SE +/- 0.05, N = 11SE +/- 0.24, N = 5SE +/- 0.29, N = 10SE +/- 0.43, N = 7SE +/- 0.35, N = 15SE +/- 0.39, N = 7SE +/- 0.14, N = 11SE +/- 0.30, N = 15SE +/- 0.29, N = 1555.5034.3042.2828.7233.5421.1652.2430.8447.5127.6351.6340.3953.7740.0050.1236.1521.9424.251. (CXX) g++ options: -O3 -lm -ldl

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 6Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P1.08652.1733.25954.3465.4325SE +/- 0.009, N = 10SE +/- 0.016, N = 9SE +/- 0.008, N = 9SE +/- 0.022, N = 9SE +/- 0.023, N = 15SE +/- 0.010, N = 8SE +/- 0.022, N = 9SE +/- 0.011, N = 9SE +/- 0.029, N = 8SE +/- 0.017, N = 9SE +/- 0.034, N = 8SE +/- 0.004, N = 8SE +/- 0.023, N = 8SE +/- 0.003, N = 8SE +/- 0.026, N = 9SE +/- 0.002, N = 9SE +/- 0.005, N = 9SE +/- 0.005, N = 10SE +/- 0.008, N = 10SE +/- 0.011, N = 10SE +/- 0.011, N = 9SE +/- 0.016, N = 92.8052.9713.0443.1693.5284.0933.7383.5503.7763.7304.4863.8724.8294.2433.3413.2512.8902.7232.8102.7563.6233.4841. (CXX) g++ options: -O3 -fPIC -lm

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e12Xeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P246810SE +/- 0.008, N = 11SE +/- 0.007, N = 8SE +/- 0.006, N = 10SE +/- 0.019, N = 8SE +/- 0.003, N = 9SE +/- 0.006, N = 6SE +/- 0.005, N = 10SE +/- 0.010, N = 7SE +/- 0.004, N = 9SE +/- 0.006, N = 7SE +/- 0.008, N = 10SE +/- 0.003, N = 8SE +/- 0.012, N = 10SE +/- 0.003, N = 8SE +/- 0.010, N = 15SE +/- 0.002, N = 11SE +/- 0.003, N = 12SE +/- 0.004, N = 11SE +/- 0.005, N = 12SE +/- 0.003, N = 11SE +/- 0.003, N = 8SE +/- 0.006, N = 82.1874.2502.3254.5763.3666.6732.6915.3323.1216.1602.3374.3242.6054.8061.3591.9211.1991.9841.2422.0584.5453.8011. (CXX) g++ options: -O3

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img CompressionXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.76351.5272.29053.0543.8175SE +/- 0.024402, N = 15SE +/- 0.023842, N = 15SE +/- 0.017240, N = 15SE +/- 0.007749, N = 10SE +/- 0.026496, N = 15SE +/- 0.048155, N = 15SE +/- 0.028863, N = 15SE +/- 0.020941, N = 15SE +/- 0.030907, N = 15SE +/- 0.023295, N = 15SE +/- 0.015943, N = 15SE +/- 0.014326, N = 10SE +/- 0.015355, N = 15SE +/- 0.012826, N = 10SE +/- 0.012053, N = 15SE +/- 0.010826, N = 15SE +/- 0.015217, N = 15SE +/- 0.015384, N = 15SE +/- 0.019468, N = 15SE +/- 0.014512, N = 15SE +/- 0.016947, N = 15SE +/- 0.023387, N = 151.4323621.8475371.4692102.1448572.1814553.3931891.7112962.2877661.9829322.6320771.7180222.2799711.8576422.5046211.0683981.3557371.1094461.3690631.1263291.4229542.3211952.2004971. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Deconvolution Batch shapes_3d - Engine: CPUXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P0.56191.12381.68572.24762.8095SE +/- 0.000902, N = 9SE +/- 0.000392, N = 9SE +/- 0.001274, N = 9SE +/- 0.000805, N = 9SE +/- 0.001091, N = 9SE +/- 0.006599, N = 9SE +/- 0.001240, N = 9SE +/- 0.000927, N = 9SE +/- 0.000806, N = 9SE +/- 0.000508, N = 9SE +/- 0.001365, N = 9SE +/- 0.002181, N = 9SE +/- 0.001467, N = 9SE +/- 0.001097, N = 9SE +/- 0.003127, N = 9SE +/- 0.001089, N = 9SE +/- 0.002070, N = 9SE +/- 0.001272, N = 9SE +/- 0.001921, N = 9SE +/- 0.001484, N = 9SE +/- 0.001685, N = 9SE +/- 0.000791, N = 90.6242500.9071610.7361701.0774300.8670651.4221300.8643581.1964300.9450751.2640702.2452602.0491002.4973402.2971200.8296530.9436211.0019141.1192301.0094731.1548701.7974901.4983501. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

CPU Power Consumption Monitor

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringXeon Platinum 8592+ 2PXeon Platinum 8592+Xeon Platinum 8490H 2PXeon Platinum 8490HXeon Platinum 8380 2PXeon Platinum 8380Xeon Max 9480 2PXeon Max 9480Xeon Max 9468 2PXeon Max 9468Xeon 6780E 2PXeon 6780EXeon 6766E 2PXeon 6766EEPYC 9754 2PEPYC 9754EPYC 9684X 2PEPYC 9684XEPYC 9654 2PEPYC 9654EPYC 8534PNEPYC 8534P170340510680850Min: 39.72 / Avg: 455.07 / Max: 946.96Min: 17.77 / Avg: 260.08 / Max: 405.42Min: 87.48 / Avg: 517.33 / Max: 758.17Min: 30.38 / Avg: 278.1 / Max: 383.24Min: 51.38 / Avg: 437.37 / Max: 583.47Min: 26.31 / Avg: 237.91 / Max: 292.14Min: 95.52 / Avg: 569.8 / Max: 760.61Min: 38.16 / Avg: 295.53 / Max: 384.53Min: 102.49 / Avg: 566.6 / Max: 756.48Min: 38.11 / Avg: 291.53 / Max: 379.76Min: 53.44 / Avg: 321.01 / Max: 713.33Min: 21.7 / Avg: 169.1 / Max: 357.69Min: 52.85 / Avg: 291.05 / Max: 542.48Min: 21.86 / Avg: 142.1 / Max: 271.22Min: 20.22 / Avg: 375.51 / Max: 735.74Min: 10.09 / Avg: 225.05 / Max: 397.44Min: 33.66 / Avg: 525.49 / Max: 837.69Min: 17.06 / Avg: 290.08 / Max: 413.51Min: 35.7 / Avg: 464.49 / Max: 751.79Min: 14.69 / Avg: 244.69 / Max: 373.1Min: 9.26 / Avg: 114.13 / Max: 155.71Min: 8.43 / Avg: 128.7 / Max: 184.62

264 Results Shown

WRF
OpenVKL
RELION
NWChem
High Performance Conjugate Gradient
TensorFlow
BRL-CAD
Xcompact3d Incompact3d
CloverLeaf
ASKAP:
  tConvolve MT - Gridding
  tConvolve MT - Degridding
PyTorch
PostgreSQL:
  100 - 1000 - Read Only - Average Latency
  100 - 1000 - Read Only
TensorFlow
OpenSSL:
  RSA4096:
    verify/s
    sign/s
Timed Linux Kernel Compilation
Stockfish
Apache IoTDB:
  800 - 100 - 800 - 100:
    Average Latency
    point/sec
PyTorch:
  CPU - 512 - ResNet-152
  CPU - 64 - ResNet-152
SecureMark
ClickHouse:
  100M Rows Hits Dataset, Third Run
  100M Rows Hits Dataset, Second Run
  100M Rows Hits Dataset, First Run / Cold Cache
Apache IoTDB:
  500 - 100 - 800 - 400:
    Average Latency
    point/sec
OpenRadioss
NAMD
Timed Gem5 Compilation
LAMMPS Molecular Dynamics Simulator
Timed Node.js Compilation
OpenSSL
TensorFlow
FFmpeg:
  libx265 - Platform
  libx265 - Video On Demand
ONNX Runtime:
  fcn-resnet101-11 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
FFmpeg
PostgreSQL:
  100 - 1000 - Read Write - Average Latency
  100 - 1000 - Read Write
Blender
Numpy Benchmark
LuxCoreRender
OpenSSL:
  AES-128-GCM
  AES-256-GCM
  ChaCha20
  SHA512
  ChaCha20-Poly1305
PyTorch
LuxCoreRender
PyTorch
Xmrig
LuxCoreRender
OSPRay
RocksDB
ONNX Runtime:
  Faster R-CNN R-50-FPN-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
OpenFOAM:
  drivaerFastback, Medium Mesh Size - Execution Time
  drivaerFastback, Medium Mesh Size - Mesh Time
Timed LLVM Compilation
Memcached
ONNX Runtime:
  ResNet50 v1-12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
NAMD
Speedb
PyTorch
ONNX Runtime:
  ArcFace ResNet-100 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  bertsquad-12 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  CaffeNet 12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
Llamafile
Timed Godot Game Engine Compilation
PyTorch
OpenRadioss
QMCPACK
ONNX Runtime:
  fcn-resnet101-11 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  CaffeNet 12-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  T5 Encoder - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Person Detection FP16 - CPU:
    ms
    FPS
ONNX Runtime:
  GPT-2 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
TensorFlow
OSPRay
ONNX Runtime:
  yolov4 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  super-resolution-10 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
OSPRay Studio:
  1 - 4K - 32 - Path Tracer - CPU
  3 - 4K - 32 - Path Tracer - CPU
Zstd Compression:
  19 - Decompression Speed
  19 - Compression Speed
Speedb
VVenC
Zstd Compression:
  19, Long Mode - Decompression Speed
  19, Long Mode - Compression Speed
OpenVINO:
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU:
    ms
    FPS
OSPRay Studio
TensorFlow:
  CPU - 1 - ResNet-50
  CPU - 64 - ResNet-50
OpenVINO:
  Machine Translation EN To DE FP16 - CPU:
    ms
    FPS
ONNX Runtime:
  ResNet50 v1-12-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Noise Suppression Poconet-Like FP16 - CPU:
    ms
    FPS
libavif avifenc
ONNX Runtime:
  T5 Encoder - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
ASKAP:
  tConvolve MPI - Gridding
  tConvolve MPI - Degridding
Helsing
OSPRay Studio:
  1 - 4K - 16 - Path Tracer - CPU
  3 - 4K - 16 - Path Tracer - CPU
CloverLeaf
OSPRay Studio
OpenVINO:
  Road Segmentation ADAS FP16-INT8 - CPU:
    ms
    FPS
ONNX Runtime:
  ArcFace ResNet-100 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Handwritten English Recognition FP16-INT8 - CPU:
    ms
    FPS
  Vehicle Detection FP16-INT8 - CPU:
    ms
    FPS
ONNX Runtime:
  Faster R-CNN R-50-FPN-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
TensorFlow
QuantLib
OpenVINO:
  Face Detection FP16-INT8 - CPU:
    ms
    FPS
OSPRay
Blender
OpenVINO:
  Person Vehicle Bike Detection FP16 - CPU:
    ms
    FPS
ONNX Runtime:
  bertsquad-12 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Person Re-Identification Retail FP16 - CPU:
    ms
    FPS
PyTorch
OpenVINO:
  Weld Porosity Detection FP16-INT8 - CPU:
    ms
    FPS
ONNX Runtime:
  GPT-2 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Face Detection Retail FP16-INT8 - CPU:
    ms
    FPS
ONNX Runtime:
  yolov4 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
GraphicsMagick
GPAW
GraphicsMagick
Timed Linux Kernel Compilation
GraphicsMagick
ONNX Runtime:
  super-resolution-10 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
GraphicsMagick
RocksDB
miniBUDE:
  OpenMP - BM2:
    Billion Interactions/s
    GFInst/s
Coremark
srsRAN Project
Blender
Aircrack-ng
WebP Image Encode
libavif avifenc
srsRAN Project
Llamafile
LULESH
7-Zip Compression:
  Decompression Rating
  Compression Rating
Algebraic Multi-Grid Benchmark
Liquid-DSP
NAS Parallel Benchmarks
ASTC Encoder
TensorFlow
miniFE
OpenRadioss
VVenC
Primesieve
GROMACS
ASTC Encoder
uvg266
John The Ripper
TensorFlow
Liquid-DSP
QuantLib
Liquid-DSP
SVT-AV1
uvg266
Liquid-DSP
TensorFlow
Liquid-DSP:
  64 - 256 - 57
  128 - 256 - 512
  256 - 256 - 512
John The Ripper
Liquid-DSP
SVT-AV1
Embree
John The Ripper
Liquid-DSP:
  256 - 256 - 57
  256 - 256 - 32
  128 - 256 - 32
Google SynthMark
Liquid-DSP
uvg266
Blender:
  Junkshop - CPU-Only
  Fishy Cat - CPU-Only
Appleseed
TensorFlow
Kvazaar
Pennant
Kvazaar
srsRAN Project
Kvazaar
OpenFOAM:
  drivaerFastback, Small Mesh Size - Execution Time
  drivaerFastback, Small Mesh Size - Mesh Time
SVT-AV1
Intel Open Image Denoise
Blender
miniBUDE:
  OpenMP - BM1:
    Billion Interactions/s
    GFInst/s
x265
Timed FFmpeg Compilation
Pennant
TensorFlow
NAS Parallel Benchmarks
PyBench
NAS Parallel Benchmarks
ASTC Encoder
NAS Parallel Benchmarks
Timed Mesa Compilation
Xcompact3d Incompact3d
Y-Cruncher
Timed ImageMagick Compilation
m-queens
Intel Open Image Denoise:
  RT.ldr_alb_nrm.3840x2160 - CPU-Only
  RT.hdr_alb_nrm.3840x2160 - CPU-Only
NAS Parallel Benchmarks
GNU Octave Benchmark
Y-Cruncher
Embree
WebP Image Encode
Embree
ASTC Encoder
SVT-AV1
libavif avifenc
ACES DGEMM
libavif avifenc
NAS Parallel Benchmarks
srsRAN Project
LAMMPS Molecular Dynamics Simulator
libavif avifenc
Primesieve
Parallel BZIP2 Compression
oneDNN
CPU Power Consumption Monitor