Intel Xeon Platinum 8592 Linux Benchmarks

Benchmarks by Michael Larabel for a future article of launch day Intel Xeon Platinum 8592 Linux review.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2312136-NE-XEONPLATI11
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:

Chess Test Suite 2 Tests
Timed Code Compilation 6 Tests
C/C++ Compiler Tests 10 Tests
CPU Massive 19 Tests
Creator Workloads 18 Tests
Cryptocurrency Benchmarks, CPU Mining Tests 2 Tests
Cryptography 3 Tests
Encoding 3 Tests
Fortran Tests 4 Tests
Game Development 5 Tests
HPC - High Performance Computing 16 Tests
Imaging 2 Tests
Common Kernel Benchmarks 2 Tests
Linear Algebra 2 Tests
Machine Learning 2 Tests
Molecular Dynamics 5 Tests
MPI Benchmarks 5 Tests
Multi-Core 36 Tests
NVIDIA GPU Compute 4 Tests
Intel oneAPI 7 Tests
OpenMPI Tests 11 Tests
Programmer / Developer System Benchmarks 8 Tests
Python Tests 7 Tests
Raytracing 4 Tests
Renderers 6 Tests
Scientific Computing 7 Tests
Server 2 Tests
Server CPU Tests 14 Tests
Video Encoding 3 Tests
Common Workstation Benchmarks 2 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

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
EPYC 8534P
December 01 2023
  10 Hours, 42 Minutes
EPYC 8534P - Power
December 02 2023
  10 Hours, 57 Minutes
EPYC 8534PN
December 06 2023
  11 Hours, 16 Minutes
EPYC 8534PN - Power
December 06 2023
  10 Hours, 55 Minutes
EPYC 9554
November 13 2023
  11 Hours, 49 Minutes
EPYC 9554 2P
November 14 2023
  11 Hours, 18 Minutes
EPYC 9654
November 15 2023
  12 Hours, 24 Minutes
EPYC 9654 2P
November 16 2023
  12 Hours, 49 Minutes
EPYC 9654 2P - Power
November 18 2023
  11 Hours, 12 Minutes
EPYC 9684X
November 08 2023
  12 Hours, 19 Minutes
EPYC 9684X 2P
November 07 2023
  12 Hours, 5 Minutes
EPYC 9684X 2P - Power
November 20 2023
  9 Hours, 15 Minutes
EPYC 9754
November 09 2023
  14 Hours, 56 Minutes
EPYC 9754 2P
November 11 2023
  14 Hours, 47 Minutes
EPYC 9754 2P - Power
November 18 2023
  13 Hours, 42 Minutes
Xeon Platinum 8280 2P
December 04 2023
  13 Hours, 16 Minutes
Xeon Platinum 8380
December 03 2023
  13 Hours, 35 Minutes
Xeon Platinum 8380 2P
December 03 2023
  10 Hours, 58 Minutes
Xeon Platinum 8490H
October 31 2023
  10 Hours, 12 Minutes
Xeon Platinum 8490H 2P
October 30 2023
  10 Hours, 46 Minutes
Xeon Platinum 8592+
December 08 2023
  10 Hours, 7 Minutes
Xeon Platinum 8592+ 2P
December 07 2023
  10 Hours, 28 Minutes
Invert Hiding All Results Option
  11 Hours, 49 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):


Intel Xeon Platinum 8592 Linux BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsNetworkMonitorOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PAMD EPYC 8534P 64-Core @ 2.30GHz (64 Cores / 128 Threads)AMD Cinnabar (RCB1009C BIOS)AMD Device 14a4192GB3201GB Micron_7450_MTFDKCC3T2TFSASPEED2 x Broadcom NetXtreme BCM5720 PCIeUbuntu 23.106.6.0-rc5-phx (x86_64)GNOME Shell 45.0X Server 1.21.1.7GCC 13.2.0ext41920x1200AMD EPYC 8534PN 64-Core @ 2.00GHz (64 Cores / 128 Threads)AMD EPYC 9554 64-Core @ 3.10GHz (64 Cores / 128 Threads)AMD Titanite_4G (RTI1007B BIOS)768GBBroadcom NetXtreme BCM5720 PCIe2 x AMD EPYC 9554 64-Core @ 3.10GHz (128 Cores / 256 Threads)1520GBAMD EPYC 9654 96-Core @ 2.40GHz (96 Cores / 192 Threads)768GB2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads)1520GBAMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads)768GB2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads)1520GBAMD EPYC 9754 128-Core @ 2.25GHz (128 Cores / 256 Threads)768GB2 x AMD EPYC 9754 128-Core @ 2.25GHz (256 Cores / 512 Threads)1520GB2 x Intel Xeon Platinum 8280 @ 4.00GHz (56 Cores / 112 Threads)GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS)Intel Sky Lake-E DMI3 Registers384GBVE2282 x Intel X722 for 1GbE + 2 x QLogic FastLinQ QL41000 10/25/40/50GbE1920x1080Intel Xeon Platinum 8380 @ 3.40GHz (40 Cores / 80 Threads)Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS)Intel Ice Lake IEH256GB2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads)512GBIntel Xeon Platinum 8490H @ 3.50GHz (60 Cores / 120 Threads)Quanta Cloud S6Q-MB-MPS (3A10.uh BIOS)Intel Device 1bce1920x12002 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)Quanta Cloud S6Q-MB-MPS (3B05.TEL4P1 BIOS)512GB2 x INTEL XEON PLATINUM 8592+ @ 3.90GHz (128 Cores / 256 Threads)1008GB3201GB Micron_7450_MTFDKCC3T2TFS + 0GB Virtual HDisk0 + 0GB Virtual HDisk1 + 0GB Virtual HDisk2 + 0GB Virtual HDisk32 x Intel X710 for 10GBASE-TOpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/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-build-config=bootstrap-lto-lean --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- EPYC 8534P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212- EPYC 8534P - Power: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212- EPYC 8534PN: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212- EPYC 8534PN - Power: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212- EPYC 9554: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9554 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9654: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9654 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9654 2P - Power: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9684X: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9684X 2P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9684X 2P - Power: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- 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 9754 2P - Power: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116- Xeon Platinum 8280 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x5003604- Xeon Platinum 8380: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0xd0003a5- Xeon Platinum 8380 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0xd0003a5- Xeon Platinum 8490H: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1- Xeon Platinum 8490H 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1- Xeon Platinum 8592+: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000161- Xeon Platinum 8592+ 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000161Java Details- OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Details- Python 3.11.6Security Details- 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 + 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 - Power: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 + 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 - Power: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 9554: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 9554 2P: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 + 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 + 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 - Power: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 + 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 2P: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 2P - Power: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 + 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 + 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 - Power: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 Platinum 8280 2P: gather_data_sampling: Mitigation of Microcode + itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + 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: Mitigation of TSX disabled- 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 + 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 + 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: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 + 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 + 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 + 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 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PLogarithmic Result OverviewPhoronix Test SuiteCpuminer-OptPrimesieveASTC EncoderminiBUDECoremarkOpenRadiossBlenderStockfishXmrigJohn The RipperStress-NGACES DGEMMOSPRay StudioHelsingrays1benchNAMDNAS Parallel Benchmarks7-Zip CompressionRodiniaEmbreeOpenVKLOSPRayOpenVINOasmFishGROMACSLULESHAlgebraic Multi-Grid BenchmarkGraph500KvazaarLiquid-DSPHigh Performance Conjugate GradientGraphicsMagickLuxCoreRenderTimed LLVM Compilationuvg266oneDNNIntel Open Image DenoiseTimed Node.js CompilationTimed Linux Kernel CompilationTachyonASKAPlibxsmmTimed Godot Game Engine CompilationRawTherapeeDuckDBTimed Gem5 CompilationSVT-AV1Timed Mesa CompilationPostgreSQL

EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PPer Watt Result OverviewPhoronix Test Suite100%274%448%622%libxsmmASKAPASTC Encoderrays1benchSVT-AV1ACES DGEMMuvg266miniBUDEKvazaar7-Zip CompressionLuxCoreRenderXmrigCoremarkCpuminer-OptGraph500GraphicsMagickNAS Parallel BenchmarksOSPRayStockfishStress-NGasmFishEmbreeJohn The RipperLiquid-DSPOpenVKLGROMACSAlgebraic Multi-Grid BenchmarkHigh Performance Conjugate GradientIntel Open Image DenoiseLULESHP.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

Intel Xeon Platinum 8592 Linux Benchmarksopenvino: Person Vehicle Bike Detection FP16 - CPUstress-ng: Vector Mathcpuminer-opt: Deepcoincpuminer-opt: Myriad-Groestlstress-ng: AVX-512 VNNIstress-ng: Vector Floating Pointxmrig: Monero - 1Mopenvino: Weld Porosity Detection FP16 - CPUcompress-7zip: Decompression Ratingopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Vehicle Detection FP16 - CPUprimesieve: 1e13cpuminer-opt: Blake-2 Sblender: Classroom - CPU-Onlyopenvino: Road Segmentation ADAS FP16 - CPUstress-ng: Memory Copyingjohn-the-ripper: Blowfishstress-ng: Wide Vector Mathjohn-the-ripper: bcryptblender: Pabellon Barcelona - CPU-Onlyminibude: OpenMP - BM2minibude: OpenMP - BM2blender: Barbershop - CPU-Onlycoremark: CoreMark Size 666 - Iterations Per Secondopenradioss: Chrysler Neon 1Mospray-studio: 1 - 4K - 16 - Path Tracer - CPUospray-studio: 2 - 4K - 16 - Path Tracer - CPUopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUospray-studio: 3 - 4K - 16 - Path Tracer - CPUstress-ng: CPU Stressstress-ng: Floating Pointblender: BMW27 - CPU-Onlyospray-studio: 1 - 4K - 32 - Path Tracer - CPUospray-studio: 2 - 4K - 32 - Path Tracer - CPUospray-studio: 3 - 4K - 32 - Path Tracer - CPUopenvino: Vehicle Detection FP16-INT8 - CPUmt-dgemm: Sustained Floating-Point Rateospray-studio: 1 - 4K - 1 - Path Tracer - CPUlibxsmm: 128stress-ng: Matrix Mathospray-studio: 3 - 4K - 1 - Path Tracer - CPUospray-studio: 2 - 4K - 1 - Path Tracer - CPUopenvino: Person Detection FP32 - CPUnpb: LU.Cblender: Fishy Cat - CPU-Onlyopenvino: Person Detection FP16 - CPUnpb: BT.Cxmrig: Wownero - 1Mhelsing: 14 digitnpb: IS.Dstress-ng: Fused Multiply-Addincompact3d: input.i3d 193 Cells Per Directionjohn-the-ripper: WPA PSKrays1bench: Large Sceneopenvino: Vehicle Detection FP16-INT8 - CPUospray: gravity_spheres_volume/dim_512/ao/real_timeospray: gravity_spheres_volume/dim_512/scivis/real_timeopenvino: Age Gender Recognition Retail 0013 FP16 - CPUospray: particle_volume/scivis/real_timeopenvino: Road Segmentation ADAS FP16-INT8 - CPUospray: particle_volume/ao/real_timeopenvino: Face Detection Retail FP16 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUnamd: ATPase Simulation - 327,506 Atomsembree: Pathtracer ISPC - Crownrodinia: OpenMP LavaMDastcenc: Mediumliquid-dsp: 240 - 256 - 512openvkl: vklBenchmarkCPU ISPCgraph500: 26embree: Pathtracer ISPC - Asian Dragonasmfish: 1024 Hash Memory, 26 Depthopenvino: Machine Translation EN To DE FP16 - CPUgromacs: MPI CPU - water_GMX50_barenpb: MG.Clulesh: amg: graph500: 26npb: CG.Cnpb: FT.Ckvazaar: Bosphorus 4K - Slowgraph500: 26uvg266: Bosphorus 4K - Slowopenvino: Handwritten English Recognition FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUkvazaar: Bosphorus 4K - Mediumonednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUluxcorerender: LuxCore Benchmark - CPUgraph500: 26graphics-magick: Sharpenluxcorerender: Orange Juice - CPUuvg266: Bosphorus 4K - Mediumhpcg: 144 144 144 - 60openvino: Age Gender Recognition Retail 0013 FP16 - CPUcompress-7zip: Compression Ratingincompact3d: X3D-benchmarking input.i3dgraphics-magick: Enhancedopenvino: Face Detection FP16-INT8 - CPUliquid-dsp: 128 - 256 - 512build-llvm: Ninjaopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16 - CPUlibxsmm: 256build-linux-kernel: allmodconfigoidn: RT.ldr_alb_nrm.3840x2160 - CPU-Onlyospray: gravity_spheres_volume/dim_512/pathtracer/real_timenwchem: C240 Buckyballopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUbuild-nodejs: Time To Compileuvg266: Bosphorus 4K - Very Fastduckdb: IMDBbuild-linux-kernel: defconfigopenvino: Face Detection FP16 - CPUtachyon: Total Timeopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Vehicle Detection FP16 - CPUaskap: tConvolve MPI - Griddingaskap: tConvolve MPI - Degriddingonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP32 - CPUopenvino: Handwritten English Recognition FP16 - CPUopenvino: Face Detection Retail FP16 - CPUopenvino: Face Detection FP16 - CPUastcenc: Exhaustiveopenvino: Face Detection FP16-INT8 - CPUcpuminer-opt: Triple SHA-256, Onecoinopenvino: Weld Porosity Detection FP16-INT8 - CPUnpb: SP.Castcenc: Thoroughbuild-godot: Time To Compilesvt-av1: Preset 8 - Bosphorus 4Krawtherapee: Total Benchmark Timebuild-gem5: Time To Compilesvt-av1: Preset 4 - Bosphorus 4Kopenvino: Face Detection Retail FP16-INT8 - CPUbuild-mesa: Time To Compilesvt-av1: Preset 13 - Bosphorus 4Ksvt-av1: Preset 12 - Bosphorus 4Kpgbench: 100 - 1000 - Read Writepgbench: 100 - 1000 - Read Write - Average Latencyduckdb: TPC-H Parquetstress-ng: Matrix 3D Mathstockfish: Total Timeluxcorerender: DLSC - CPUjohn-the-ripper: MD5EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2942.16371417.8619527263905505982.93135537.6220536.721.57385839328.9322.3242.83933657764.8154.7816465.121055871667888.5910583180.712806.348112.254236.612415599.084926295.4937050371159775.441046.3842664132960.6918641.7526.016767167751790833123.3922.82206719051304.9266664.8422611922144.93137060.0833.14145.25137005.0139364.094.4222757.07124404567.2220.9124565402229341.2810.1316.706316.145565468.5116.589830.4816.63484.5411.220.3817465.658145.380284.6510861733333132633687700080.406115792063897.086.26063349.2916684.800115278333326794500022438.7268083.6224.1675659000017.6342.9610.7924.731.504197.8472333200053014.5519.9322.24060.72354386560.360494902566.57844210000179.54151.542951.491818.0330.0201.9118.93501955.480817.290.53165.10152.36123.60543.65729.527.6589583.021427.131666016838.50.5198051239.06220.00220.491485.626884.011080.153.885256.282350475623.0152624.7137.9074137.46790.51859.455222.7574.8256.4124.331186.038186.1246191616.153171.6596004.421769623959.6795150003288.75420129.7621470299176602897.65154125.2120805.718.41430589367.7922.4333.97136428054.4555.0718906.331141122018031.7211457268.213167.597126.704203.642593724.278507296.41327293308911615.141168.4837936145093.0519966.2721.765936060104692503503.9123.05186416571321.7311884.1319731678146.47140067.0828.27147.42137649.2242491.178.9372798.20149843902.8120.5405852418289406.069.0117.916017.450873571.8117.255427.2917.27603.919.290.3316073.827438.185313.92421031033333141934183700086.935218045365586.846.74564451.0216474.640115474066726923300022681.8268882.4724.2377961000018.0538.219.6524.881.337208.6076272800056316.1720.4822.32280.64364785555.297994968475.331004166667168.03043.263455.342028.4295.2512.1320.99511742.590317.600.48157.53953.66123.35741.33034.2123.2283580.021419.7116726.716838.50.4623581475.85216.77218.191668.827963.88930.824.451867.082689436751.6753199.7242.1052131.91991.66759.475217.5184.8105.3623.551187.758187.5056224416.068171.7336000.6820866444211.33108853332545.14328890.3617353233704688659.48119077.5820394.024.59343521295.7922.3354.28329979375.1456.0314538.42942661406939.129422993.332487.44399.498267.692152972.955463301.1740962411338305.17938.5047070118644.6216586.4330.097550775559877892813.7021.14679121451244.9233716.2325462163162.65132191.0238.08162.25133857.4335279.6113.1392707.25106714867.6720.8184172340741270.5411.2614.951214.484258065.1114.837034.0014.86095.4612.970.4299559.451152.115245.9950751586667119831866800075.3136144229352107.985.71863818.0916431.062115066300024946300022464.4367161.3922.8269394500016.3147.6912.4823.411.706417.0165662900048413.2018.3722.24920.82332890561.969991834645.20741760000191.73358.792590.451621.6348.4821.7116.93382189.171656.630.61174.55248.21126.95545.03125.9832.0899570.011426.1916190.416066.50.6158841086.60196.89196.441338.895697.821226.113.415649.422075304878.4053144.2132.4111142.98984.60860.315231.4944.7857.5225.038178.673182.1455717817.491172.1965986.411535975748.6484056672609.31389232.7020300275005221186.42139448.4920455.423.38409338310.3922.1945.76834615066.5656.6416738.591121151602887.5211156584.312976.003119.040241.012484026.786375296.4635655359428850.90971.6841225135689.2519466.4626.516521666039762972914.8622.86263318331305.5259675.5021851857148.15139476.2034.80148.14137749.2540241.8103.6892774.41118913171.7620.7874679359639306.5210.8617.308316.717259330.5717.152732.8417.20045.3211.930.3854468.600646.107265.0553937776667135835072700083.8198162607206102.916.08363431.6416508.772115258266727465700022415.0468641.9223.8377408700017.6843.9412.1824.421.537438.2375628900056314.819.9622.30800.81357061556.880900951597.72923553333173.85651.932723.421653.8314.9011.8719.84841926.775320.930.59160.06852.88125.89742.30127.8228.9170563.771435.3316928.816750.30.5790511229.83215.71215.691452.185852.151144.573.761353.332444855291.7053304.5235.0884133.94689.28059.620218.0624.8047.0924.023186.727184.3466186816.163171.4945994.3617213283410.1087356674260.64507522.1625933360837240174.36181377.9747926.415.30521569465.1412.9728.13643478047.2424.2222827.201380322250265.2713803557.543883.576155.343170.563056532.141324204.87220832235613514.521550.4931236116427.0424168.3618.714969949895579604691.8830.17490414012400.4346518.0216711412105.91253193.8423.9994.04233622.4460032.471.2685184.36167642337.909.74551621512773469.776.7221.312720.739087092.8121.043220.5821.12533.227.890.2822789.688734.014355.183511760000001712467453000107.874821016705468.679.651125921.2924037.442232115600036801600045867.44124827.7529.2186515200021.7629.747.4429.991.1521110.4883615900066918.5224.6422.45110.54489736515.7927581142402.091145233333131.20335.564151.232524.6238.2892.6424.73391452.1107563.740.4121.92863.52103.72033.51641.7920.32301317.722458.3437265.931177.20.3850511795.33339.80302.062143.499442.97763.585.025279.473137637920.45119472.3047.6034103.500107.90449.943179.7435.8144.5619.398201.391200.8547286013.726143.30613521.3424177432813.04131793337329.87990315.62509307089314385828.77350580.6782864.515.74947599820.215.2214.67788120025.2312.4745325.212437234381848.8324711531.147369.796294.79296.445801339.80844685.72122901248126807.792853.5614599271343.9647698.1410.282650528937345738978.4554.7389437734031.3688593.4991678547.93476679.4913.5647.89452577.1398348.138.1939783.65333205139.824.87686817950002803.653.5537.231836.0232146208.6737.137811.2037.24311.758.010.18644153.866620.545514.766621022000002806670675000176.209037049328338.9616.941225076.9246784.684454522933349649200085050.58243408.8344.62130512000033.6132.364.3645.291.2126210.471258070000103826.9037.5049.00620.51734093235.7495861600206.80171636666790.32738.318037.054643.4157.5944.1830.19501322.7157704.810.3493.41356.40132.29226.81182.6210.77932562.826103.9273231.661441.90.3094313333.34667.62667.113944.9218148.52386.609.5575154.4862115215810.50242455.0790.075492.979101.29658.333156.2055.7484.7518.455181.699176.6946035216.570148.51519439.8941258651116.79221114004910.14611736.5931113435938271540.06215564.7859047.719.09648630561.9818.0924.16454170739.9338.2528310.821668872597630.0516685048.774400.504176.020142.383884296.768928176.28182651843516149.991879.9521550220744.8830824.9715.894194242520485055665.0839.28215811422578.6421053.0213481149153.11271216.0420.32150.31261864.1175127.461.5295195.24202941117.458.45212936649345541.158.3426.149825.508288979.2124.889425.4724.94854.109.830.26536111.033830.240407.246414002000002099499347000132.841025020902085.2511.408118608.0723701.442229673066739116700054465.04125164.0339.4683991700028.5536.689.6940.071.2269811.6081617200080921.1132.0132.12830.80564950430.0803291319500.741172100000133.07044.564989.002858.8284.2492.9827.08651568.4109112.780.6120.15663.46115.80536.60050.2617.16941262.822645.5747003.241081.60.3814132148.82321.55313.532608.9411192.67951.065.960595.653809279559.37125373.8756.4061101.563110.29152.051180.8575.8725.7719.921190.126193.2557821212.786146.27613448.2028906690914.48157630009086.081197434.30622438646016300097.80423319.18103949.618.9912296861000.316.3512.528107278320.8715.5455966.753115215107325.6630820725.708300.420332.01782.488035017.41506571.48102751043132601.913488.9212135326382.1860101.698.6520581207382776610933.9469.6713336444275.8816181.0376565360.23497487.0011.5360.42509783.05125526.031.66010049.34399961883.834.293908981222333893.474.3743.751642.5493135374.6342.384513.7442.37612.159.790.12938179.826418.239464.464122591333333338857107000212.934737233515547.9318.962193302.5545810.905451477033358196500085652.95231864.8761.24154399000044.6939.105.2762.961.4652110.481456580000112227.6446.7943.97010.54809868247.1784821660250.24161916666791.01046.319948.895647.3188.5704.3929.95271559.5158898.750.3493.05154.69154.97128.860102.599.24113085.227541.3185111.777438.70.3570814138.94794.35796.464899.5722196.22466.9411.7036191.3374981319365.67263543.40105.483892.427109.69564.678162.8675.8075.8618.961187.847190.3545717417.492156.69618432.2951136245219.09271533339702.811370952.74721139820720096149.61486342.09112376.517.4913652121055.575.9311.673122832018.1413.9562243.073520916004812.7937881222.739152.759366.11172.208770195.33610667.668742885136200.373994.6510364474925.6067867.727.6217500177872075011978.8079.8155285474749.7965326.5864855450.78524075.1110.5651.40551536.08136083.228.8669969.47453551958.374.193066021337333965.943.9951.620049.8644153305.2149.800312.0049.80652.008.960.11248205.225616.435554.120627066666673850948166000235.414940876460945.4121.082195724.0246825.421454912566765602500085165.37232777.4863.18172980000044.6736.154.9463.901.4943810.581617340000133433.5345.2250.86250.52861035191.0233491934229.66168806666785.06141.7310749.676053.2174.2384.6832.05631440.1157993.590.3489.41453.01152.22828.119108.358.02203435.768081.0410157387476.50.3397654592.13933.66944.575301.9623914.31441.6313.6902208.8286981321326.43272752.37120.032490.881109.36364.928153.2715.8065.2818.819190.762189.5665758917.368157.76519863.3053648902622.05287100005238.23585746.5630810428138459429.38214129.8569205.620.26623489550.9612.2125.96952330340.5227.7726495.631660802597371.5916611949.844251.197170.048142.543703841.71709486.98181911826914961.491915.1621408211280.2329220.7816.284218341916484505425.6540.32006211323102.4417406.3913351142106.90333543.8120.69100.51312135.7373567.364.3496138.54202387450.787.30814419611601519.388.7326.415225.748884074.2225.040125.0025.05144.3210.560.26591117.461630.317398.983013381666672145501222000143.097923392663186.9711.546137608.6824712.884237485333338375400059557.08124391.0038.6388218100027.6037.489.0939.231.2369711.3185402900077521.2330.9423.77840.85554411440.5485841277529.301142300000131.99046.414704.823147.8283.8652.4526.75301641.6101207.730.65119.20360.62116.05036.59148.0217.93361722.743903.8267965.558310.10.3978692064.47476.94448.722553.2410693.40997.246.051490.463729938926.33208538.3256.473098.555107.89552.991180.0615.8246.1720.082180.678181.3706760614.792147.48916634.4428198271114.06147996678754.431146015.75613478474016651814.29419680.33112646.920.021203430981.626.4113.294104971021.2415.4951865.123238555100138.7130909126.607890.082315.60377.937647272.54634653.5399541004931334.583436.3911680366232.6157545.538.7819903199712328410367.4869.3986196235420.6817035.0773262848.40516601.6611.6348.61625624.70119142.632.67311652.16393382814.534.023657851137168863.844.6144.211742.9636122698.6742.100313.9542.15532.3110.530.12889201.389118.318478.234021664000003534825625000233.662936285945648.8218.775219072.0447999.677465930566758978500081967.39229308.7761.94148449000041.9438.445.4761.581.5860010.271406670000108926.3543.2944.11220.55802007194.7439021646269.7153573333391.74548.019459.515893.2186.8533.7729.00971614.6145836.620.3692.35555.22154.19229.08395.539.44143097.027470.141349601124790.3892513995.51986.68991.144966.2420657.27501.2311.9087177.5574135718004.36353106.44104.877989.617106.91865.000159.5825.7706.119.165172.712175.7165465718.296158.48422460.1947946263122.73244117339614.161339784.62706479596019644452.75475894.31123488.317.7513492171113.295.7511.885120310318.4613.6961028.673773185929288.2337390623.338928.991357.16071.098618342.53043646.958624869236106.604022.3410240463915.9466261.187.8717213173962032411860.8578.9815905407125.3939452.3763954843.41569289.3610.5943.37708202.58131097.729.33412345.62442132595.981302333952.844.0451.125949.5366142203.3148.546811.9248.57252.049.080.11299226.873616.665518.791826579666673964885000000262.204239850737943.0621.107235405.2849234.585467521633364507600087251.56242247.8562.39159545000045.9333.764.9863.221.5272310.671487900000131130.4247.5373.26100.538534271931232.52169486666785.45041.6610682.546303.0169.8794.1032.3982157084.790.3487.74555.67154.75327.868108.528.18033503.218334.111476151268070.3550314605.881105.951105.045660.3823435.35441.2113.4865206.0685238020993.09390558.06116.987687.169109.83464.358157.5245.8295.2918.781178.470178.4865598617.870156.15724999.4253747317221.49287643336030.81773130.23410305552311411104.74280779.0229356.120.86774122537.1242.5921.75670588331.6496.4134497.512046773437154.9620482839.535892.337235.694118.444543138.116767212.08147521494820080.541708.7717551278579.2339031.6412.993606636136415465762.2743.6819239242702.5552067.041097934257.82279880.9116.84251.05296367.0477607.248.9475773.25256387572.729.02806168779799587.3811.0132.318731.091191245.0230.726337.5330.82414.4910.710.22836122.188625.150456.771617147000002490502151000151.4150290260505119.1312.202127404.1622356.746229104966737703300044431.58141255.0738.14118451000029.2451.0210.5338.811.2932812.43114709000092424.6632.6825.89181.07586571493.5024521451542.601384533333146.38648.646088.853343.5319.4263.5928.61431700.8112191.740.76130.46055.11147.60141.54561.013.2615668.591502.8227269.226593.40.3848442624.94258.36248.432504.0913857.821043.508.0402117.4649515011729.45131257.1574.3537118.24797.97566.132208.5764.9436.2023.589177.549178.5987148313.989177.1348009.2136332964216.76144566679815.911512021.208089011105322341175.74546565.0685822.020.6014620161123.969.6211.173139156016.5120.8167296.574126026653978.0042197721.757870.834314.83368.918228175.70286190.058205832939496.624202.089762495158.0475275.597.3216512167021954213075.2786.3311705173982.01062616.6261652685.20596937.7810.1884.31507704.84138322.728.08311222.70497503915.394.413823251481533998.924.8854.136952.6320129173.9049.494915.2149.61642.3210.820.10703206.611916.455475.9879263676666741081074790000249.844537575326856.8819.818246953.1943781.190453175233373279900066627.47243422.1755.03204045000038.0345.386.5054.651.838509.631865120000124233.6038.6147.00530.64827103249.0711671805271.81141833333398.89850.1712118.396236.9205.9974.6728.12581721.4143642.660.43104.59044.27202.19133.520121.937.17333074.676642.5268286.466643.00.4104655097.39762.57750.985633.7427396.27523.9815.8161235.0099458023307.15218198.49130.6379110.07098.96281.019185.7534.8996.4222.339179.232178.5165755717.391191.65217903.5858115532621.243133333310360.981624188.718540011886024419169.14587668.3886410.119.6016043081292.019.759.844145250015.2316.9673195.744351937475976.3342849419.927970.166318.80762.688229554.41901973.777397751742407.404577.358837557160.7477279.846.7114814150841766213902.6688.1313744624190.81164762.8655447083.88616796.989.4282.03514956.56144416.426.70311885.21554322523.2114970001077.624.5956.388555.0268133758.9350.154113.9750.23492.219.970.09758218.989915.526393.3814264733333343051086280000258.649838519765249.4821.902249368.2544011.654454989466777797900067935.26249256.0655.95202988000038.1540.576.1654.921.828059.931835480000132832.1638.9051.74740.638509831879251.33141613333395.90746.5612835.185875.2212.9844.8928.2069147460.160.43103.87546.13202.06933.854131.846.91143769.746561.3969975.467898.50.4027105491.84785.45763.106299.7928813.74483.4516.7606254.05104796325362.34224196.76138.3845110.09799.47781.331182.7504.9215.9922.467177.464175.5615662617.661190.63818848.5462278170120.43337599331213.19190997.9414413199633222147.3293651.6516581.330.32241284209.2020.6054.52726871797.2742.8112751.22776701466592.9178214115.901556.27962.251349.231569462.317751257.8447446482878143.60767.6655097109916.2711247.8834.4985585876231012352759.7619.39221524171431.8258693.3829062467133.91115269.9945.23133.20134960.2427482.9143.6332060.06112663200.8318.4491412375917238.535.0012.248212.059848898.5910.726418.2110.67694.268.210.3786246.034259.652151.585273881000099422860400059.218613480975866.865.82367474.8618165.240149553533317997700032665.5758881.1415.5749103100011.25101.0111.4916.035.620233.1548058700052810.6512.8928.97641.13291900444.934754803204.74710450000218.51490.371839.12450.5353.8812.1714.24622686.6106600.220.46196.61230.09136.16241.50218.64128.8550326.68677.1811749.210018.813.14004619.26105.01104.45553.913239.91748.402.111368.311422036817.3149939.5618.9054160.89559.98859.593235.8653.4876.7328.341135.315135.8305671517.632177.7606902.711242911346.5475731331103.97173931.239196.71132072728929.1967329.9714453.831.2193509156.0536.4380.291180420122.1188.539537.97570111136667.1156989148.861243.41549.737459.671200098.069534342.7452938535165978.06646.036198280158.1111125.8545.841002671012611171482125.0613.62280629831110.7181695.0835212982238.4198350.9858.66238.89115701.6323745.6161.8922333.9892550459.3324.0121949252076182.329.219.899269.7141827152.258.9353730.859.018928.378.510.5304243.781677.083111.939055682000090030041100056.057688111984127.914.84356195.1818008.014107927700022143700022768.3659708.3515.2080265900011.0682.0118.0215.586.639014.737875450003518.9812.8220.65441.32248249551.260030612413.59545333333264.126110.021267.60300.4451.8541.8011.86523505.460464.310.55229.91230.57110.72947.55812.3647.9070225.49545.339965.288945.964.00909398.6083.5383.70485.282357.531604.961.488348.24941534520.0137550.9613.5947169.29574.68753.384251.6314.0636.4828.450165.204165.0566476915.448167.2236436.61904652435.6054756671994.06339670.8318007258775531509.89132436.9425209.032.73331094283.1117.6340.80834539362.2538.5918459.481124582227716.1911321676.312386.40495.456239.182281894.055764182.96301193037413525.451194.8634590153923.0922002.8223.625458654988636984427.0024.96677215282006.1353381.3218031540114.61185826.4530.86113.75192791.1241014.484.5163338.68181639452.0411.0611921488110333.974.5018.977518.441260574.4016.393816.7116.44454.218.210.2811582.669844.552211.28191066900000155639089300099.752516386041370.519.047113810.7233249.861211994466729329900039597.21100227.4819.7590414100013.9580.269.9920.383.952036.2887443000063914.2416.3040.31231.29411957289.8966681077206.28969276667159.17770.902425.37600.2253.7993.0220.64052198.3125095.630.44149.18041.78124.30833.68824.6224.7158517.741131.5219879.217571.32.095681128.84175.66174.35996.054723.40808.462.973096.821843009727.1590836.7826.8557131.14379.18355.527193.0214.0895.9024.964172.122172.8767162413.962169.56712793.841725109989.0597097503543.19315559.6216535244014947898.66138052.1927608.56.43299902462.9815.8952.37526778369.2940.1018809.55939134423715.199387788.512146.28885.851273.442233945.606722205.8631047312349340.551098.3135948146387.6122454.7025.915720357985674213231.6623.25738816451764.6304345.3019651662103.33136342.9135.65103.46180757.4335870.286.7582885.71145542113.6612.7983420399141334.7718.3719.358518.808687274.9519.917154.4819.95099.653.510.2881268.582242.713298.49981013766667174045448300091.247113081859464.608.64588049.9023997.715161182600033312900035560.2570989.9720.26111312000015.8238.6816.8321.020.2500147.09107386000069113.0917.9331.24320.61389690353.5575951124361.56905980000169.52848.769096.261030.1289.9662.6920.08432056.6105652.290.48151.56749.7599.24935.60277.8125.0086746.671874.9422365.219616.00.3113361224.63289.65290.081540.076108.87384.694.1094165.4919458316507.1372703.3542.3293128.46790.06646.578197.7594.9306.2822.670194.179194.6237060414.164148.0999894.201534692658.5295773337656.87551768.9032390466979195364.46251128.0545484.34.37547633743.367.6827.42253483736.6619.5733544.081781127733151.5917810847.163544.568141.783146.554492941.732419114.30140431423221058.401909.2716737260500.5939566.7013.963364534029390555793.7945.7867988862948.0535739.58105990646.85263384.9219.7046.93336631.6960698.350.4843954.75254737912.806.17124205766122611.8520.6336.512435.2898141769.5734.667962.7934.64338.032.520.24655123.299526.661487.782714175666672685521570000152.485523503260840.1415.221162844.2345133.446317453866737965500061041.15118612.3331.64109422000025.0627.2715.5932.500.4333227.981051290000117317.1828.0160.42240.38604131193.5731301763288.821209600000110.86139.9927014.311575.9183.8914.4614.88401879.7150083.470.29108.09554.85123.83327.687217.1112.97601531.593900.2843896.4382130.3343532998.94638.73639.864392.7314919.93138.048.0607414.7738518347285.19152640.6579.1489109.56180.62153.671161.4394.7705.6820.672187.321185.9946756514.801156.79018308.8427537160110.88175097505333.79340324.7017580267037327410.00147013.6040381.23.84308961567.756.5449.06228585765.0216.2020486.971042324735239.9910406781.162682.324107.293244.982410201.883069157.17228602298012358.741252.2331415156861.8923555.9223.965007050242584774694.2229.13678714271873.4301894.531691143040.88233511.6232.7040.99236490.8642867.684.9483974.93159324513.2910.1814694428710378.2713.6222.846022.413472499.6823.583851.0723.60656.852.370.2536481.792139.886323.340211804333332012547915000111.883815400340928.1010.504107735.1239468.910183991266741913800066065.9698939.8624.33130420000015.9124.4111.9924.750.2149607.88123867000074914.4716.7235.42140.52448786323.5274151192232.131146933333148.73538.2116646.011293.1245.0793.1022.65841850.880814.500.44134.09335.8996.87030.767121.1922.8482986.422440.0537712.330422.70.2832501673.95389.78390.862604.009331.06131.894.5637275.4620883326776.45111687.4846.4627111.96096.99445.530174.1755.6355.1720.010210.900209.0967523213.293134.73213854.381675942909.40995626710169.60625097.39359175216313187369.52264466.3770136.23.936276061101.846.6425.13157095734.7017.0036840.381895368548723.3619639843.304875.547195.022153.005001821.74618184.09120791221524813.702388.2014106280684.8643220.2012.702384524038330708918.8658.6428497483209.0536146.5888775142.55462410.5617.5742.54437263.7474961.149.3177849.17292027758.585.24690800797359660.9814.3442.907741.7054115238.2441.738453.5641.81776.972.440.18084151.772024.778536.084320640333333280653964000199.147327215535628.9718.343217625.3272625.296373948533347627400096083.42181631.9241.51124833000027.4824.8612.5542.360.3713538.501199830000130219.5429.9570.95240.46670551167.0851591936237.36144163333396.46938.5632258.522274.5150.7204.5116.98321738.6122410.810.3998.32856.04124.71323.776237.8311.95391880.124799.0272808.359984.30.2962343317.93751.31751.245110.0018346.00134.418.8528538.2041393749016.72242574.3887.946493.30284.35157.793146.6235.3515.1518.230197.088193.1066385215.663149.98025487.4229753507212.4119053000OpenBenchmarking.org

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2K4K6K8K10KSE +/- 3.42, N = 3SE +/- 3.88, N = 3SE +/- 4.28, N = 3SE +/- 4.75, N = 3SE +/- 7.61, N = 3SE +/- 8.48, N = 3SE +/- 1.52, N = 3SE +/- 4.12, N = 3SE +/- 16.63, N = 3SE +/- 1.37, N = 3SE +/- 9.17, N = 3SE +/- 8.70, N = 3SE +/- 5.40, N = 3SE +/- 13.35, N = 3SE +/- 5.71, N = 3SE +/- 1.87, N = 3SE +/- 0.73, N = 3SE +/- 0.28, N = 3SE +/- 0.29, N = 3SE +/- 6.84, N = 3SE +/- 0.98, N = 3SE +/- 11.25, N = 32942.163288.752545.142609.314260.647329.874910.149086.089702.815238.238754.439614.166030.819815.9110360.981213.191103.971994.063543.197656.875333.7910169.601. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector MathEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P300K600K900K1200K1500KSE +/- 2.39, N = 3SE +/- 35.15, N = 3SE +/- 23.96, N = 3SE +/- 287.38, N = 3SE +/- 91.37, N = 3SE +/- 103.25, N = 3SE +/- 84.27, N = 3SE +/- 82.56, N = 3SE +/- 131.23, N = 3SE +/- 77.65, N = 3SE +/- 125.17, N = 3SE +/- 3034.56, N = 3SE +/- 1066.79, N = 3SE +/- 311.95, N = 3SE +/- 784.30, N = 3SE +/- 329.15, N = 3SE +/- 197.06, N = 3SE +/- 148.92, N = 3SE +/- 127.07, N = 3SE +/- 115.63, N = 3SE +/- 199.41, N = 3SE +/- 1109.88, N = 3371417.86420129.76328890.36389232.70507522.16990315.62611736.591197434.301370952.74585746.561146015.751339784.62773130.231512021.201624188.71190997.94173931.23339670.83315559.62551768.90340324.70625097.391. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: DeepcoinEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20K40K60K80K100KSE +/- 3.33, N = 3SE +/- 70.24, N = 3SE +/- 16.67, N = 3SE +/- 34.64, N = 3SE +/- 12.02, N = 3SE +/- 10.00, N = 3SE +/- 49.10, N = 3SE +/- 12.02, N = 3SE +/- 40.96, N = 3SE +/- 10.00, N = 3SE +/- 14.53, N = 3SE +/- 49.10, N = 3SE +/- 11.55, N = 3SE +/- 72.11, N = 3SE +/- 141.89, N = 3SE +/- 43.33, N = 3SE +/- 6.55, N = 3SE +/- 17.64, N = 3SE +/- 185.00, N = 4SE +/- 10.00, N = 3SE +/- 5.77, N = 3SE +/- 377.90, N = 319527.0021470.0017353.0020300.0025933.0050930.0031113.0062243.0072113.0030810.0061347.0070647.0041030.0080890.0085400.0014413.009196.7118007.0016535.0032390.0017580.0035917.001. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: Myriad-GroestlEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 10.00, N = 3SE +/- 6.67, N = 3SE +/- 40.00, N = 3SE +/- 46.19, N = 3SE +/- 3.33, N = 3SE +/- 80.07, N = 3SE +/- 20.28, N = 3SE +/- 231.80, N = 3SE +/- 68.39, N = 3SE +/- 130.43, N = 3SE +/- 15.28, N = 3SE +/- 110.60, N = 3SE +/- 12.02, N = 3SE +/- 29.06, N = 3SE +/- 134.54, N = 3SE +/- 14.53, N = 3SE +/- 48.07, N = 3SE +/- 33.83, N = 3SE +/- 213.63, N = 15SE +/- 356.71, N = 3SE +/- 6.67, N = 3SE +/- 63.33, N = 326390299172337027500360837089343593864609820742813847409596055523111053118860199631320725877244014669726703521631. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: AVX-512 VNNIEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P5M10M15M20M25MSE +/- 927.39, N = 3SE +/- 710.35, N = 3SE +/- 4958.83, N = 3SE +/- 8358.41, N = 3SE +/- 3784.35, N = 3SE +/- 10022.95, N = 3SE +/- 527.77, N = 3SE +/- 11868.54, N = 3SE +/- 20573.29, N = 3SE +/- 7626.35, N = 3SE +/- 19986.49, N = 3SE +/- 29826.06, N = 3SE +/- 6288.41, N = 3SE +/- 918.94, N = 3SE +/- 39889.18, N = 3SE +/- 7024.85, N = 3SE +/- 26149.18, N = 15SE +/- 5843.21, N = 3SE +/- 9446.68, N = 3SE +/- 29469.24, N = 3SE +/- 51700.38, N = 12SE +/- 60867.75, N = 35505982.936602897.654688659.485221186.427240174.3614385828.778271540.0616300097.8020096149.618459429.3816651814.2919644452.7511411104.7422341175.7424419169.143222147.322728929.195531509.894947898.669195364.467327410.0013187369.521. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector Floating PointEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P130K260K390K520K650KSE +/- 313.29, N = 3SE +/- 637.76, N = 3SE +/- 158.26, N = 3SE +/- 210.33, N = 3SE +/- 501.85, N = 3SE +/- 389.19, N = 3SE +/- 711.78, N = 3SE +/- 459.74, N = 3SE +/- 899.32, N = 3SE +/- 391.96, N = 3SE +/- 843.25, N = 3SE +/- 1347.54, N = 3SE +/- 685.71, N = 3SE +/- 2025.56, N = 3SE +/- 1577.10, N = 3SE +/- 24.19, N = 3SE +/- 159.50, N = 3SE +/- 209.06, N = 3SE +/- 194.19, N = 3SE +/- 1541.65, N = 3SE +/- 350.99, N = 3SE +/- 2705.65, N = 3135537.62154125.21119077.58139448.49181377.97350580.67215564.78423319.18486342.09214129.85419680.33475894.31280779.02546565.06587668.3893651.6567329.97132436.94138052.19251128.05147013.60264466.371. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

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 Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1MEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 7.92, N = 3SE +/- 9.40, N = 3SE +/- 3.26, N = 3SE +/- 241.93, N = 3SE +/- 42.25, N = 3SE +/- 260.16, N = 4SE +/- 142.78, N = 3SE +/- 272.28, N = 4SE +/- 221.70, N = 5SE +/- 80.10, N = 4SE +/- 327.17, N = 5SE +/- 179.21, N = 5SE +/- 2550.91, N = 12SE +/- 918.55, N = 5SE +/- 800.08, N = 4SE +/- 136.51, N = 15SE +/- 35.48, N = 3SE +/- 227.50, N = 7SE +/- 7.51, N = 3SE +/- 285.92, N = 3SE +/- 12.73, N = 3SE +/- 46.27, N = 420536.720805.720394.020455.447926.482864.559047.7103949.6112376.569205.6112646.9123488.329356.185822.086410.116581.314453.825209.027608.545484.340381.270136.21. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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 2023.2.devModel: Weld Porosity Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.16, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 321.5718.4124.5923.3815.3015.7419.0918.9917.4920.2620.0217.7520.8620.6019.6030.3231.2032.736.434.373.843.931. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 RatingEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P300K600K900K1200K1500KSE +/- 651.19, N = 3SE +/- 539.70, N = 3SE +/- 173.67, N = 3SE +/- 874.48, N = 3SE +/- 599.23, N = 3SE +/- 5312.23, N = 3SE +/- 609.70, N = 3SE +/- 1320.25, N = 3SE +/- 2048.52, N = 3SE +/- 3771.78, N = 3SE +/- 1760.30, N = 3SE +/- 4828.11, N = 3SE +/- 957.95, N = 3SE +/- 15822.85, N = 3SE +/- 1731.41, N = 3SE +/- 1176.01, N = 3SE +/- 70.59, N = 3SE +/- 2725.46, N = 3SE +/- 128.86, N = 3SE +/- 19990.49, N = 3SE +/- 1471.75, N = 3SE +/- 750.30, N = 33858394305893435214093385215699475996486301229686136521262348912034301349217774122146201616043082412841935093310942999025476333089616276061. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Machine Translation EN To DE FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30060090012001500SE +/- 0.33, N = 3SE +/- 0.42, N = 3SE +/- 0.48, N = 3SE +/- 1.05, N = 3SE +/- 0.10, N = 3SE +/- 0.24, N = 3SE +/- 0.64, N = 3SE +/- 4.80, N = 3SE +/- 2.28, N = 3SE +/- 1.04, N = 3SE +/- 1.73, N = 3SE +/- 3.01, N = 3SE +/- 6.47, N = 12SE +/- 9.78, N = 3SE +/- 2.96, N = 3SE +/- 0.19, N = 3SE +/- 1.01, N = 3SE +/- 0.53, N = 3SE +/- 0.24, N = 3SE +/- 2.33, N = 3SE +/- 3.70, N = 3SE +/- 3.11, N = 3328.93367.79295.79310.39465.14820.21561.981000.311055.57550.96981.621113.29537.121123.961292.01209.20156.05283.11462.98743.36567.751101.841. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.14, N = 14SE +/- 0.00, N = 3SE +/- 0.21, N = 14SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.62, N = 13SE +/- 0.11, N = 4SE +/- 0.08, N = 15SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 322.3222.4322.3322.1912.975.2218.096.355.9312.216.415.7542.599.629.7520.6036.4317.6315.897.686.546.641. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 8.0Length: 1e13EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.019, N = 3SE +/- 0.076, N = 3SE +/- 0.009, N = 3SE +/- 0.131, N = 3SE +/- 0.030, N = 3SE +/- 0.024, N = 4SE +/- 0.015, N = 3SE +/- 0.020, N = 4SE +/- 0.021, N = 4SE +/- 0.020, N = 3SE +/- 0.007, N = 4SE +/- 0.013, N = 4SE +/- 0.014, N = 3SE +/- 0.017, N = 5SE +/- 0.020, N = 5SE +/- 0.054, N = 3SE +/- 0.410, N = 3SE +/- 0.105, N = 3SE +/- 0.073, N = 3SE +/- 0.044, N = 3SE +/- 0.041, N = 3SE +/- 0.031, N = 342.83933.97154.28345.76828.13614.67724.16412.52811.67325.96913.29411.88521.75611.1739.84454.52780.29140.80852.37527.42249.06225.1311. (CXX) g++ options: -O3

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: Blake-2 SEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P300K600K900K1200K1500KSE +/- 42.56, N = 3SE +/- 150.11, N = 3SE +/- 12.02, N = 3SE +/- 100.17, N = 3SE +/- 5.77, N = 3SE +/- 475.01, N = 3SE +/- 1481.69, N = 3SE +/- 6984.26, N = 3SE +/- 1292.56, N = 3SE +/- 21.86, N = 3SE +/- 43.59, N = 3SE +/- 1962.03, N = 3SE +/- 118.65, N = 3SE +/- 840.32, N = 3SE +/- 100.17, N = 3SE +/- 134.21, N = 3SE +/- 1004.81, N = 3SE +/- 722.36, N = 3SE +/- 6.67, N = 3SE +/- 8.82, N = 3SE +/- 24.04, N = 3SE +/- 37.56, N = 33365773642802997933461504347808812005417071072783122832052330310497101203103705883139156014525002687171804203453932677835348372858575709571. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Classroom - Compute: CPU-OnlyEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.21, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 4SE +/- 0.12, N = 3SE +/- 0.19, N = 3SE +/- 0.09, N = 3SE +/- 0.17, N = 3SE +/- 0.05, N = 3SE +/- 0.45, N = 3SE +/- 0.49, N = 364.8154.4575.1466.5647.2425.2339.9320.8718.1440.5221.2418.4631.6416.5115.2397.27122.1162.2569.2936.6665.0234.70

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 2023.2.devModel: Road Segmentation ADAS FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.25, N = 3SE +/- 0.07, N = 3SE +/- 0.35, N = 3SE +/- 0.12, N = 3SE +/- 0.25, N = 3SE +/- 0.01, N = 3SE +/- 0.87, N = 15SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 2.03, N = 15SE +/- 0.19, N = 14SE +/- 0.07, N = 3SE +/- 0.36, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 354.7855.0756.0356.6424.2212.4738.2515.5413.9527.7715.4913.6996.4120.8116.9642.8188.5338.5940.1019.5716.2017.001. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Memory CopyingEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P16K32K48K64K80KSE +/- 4.27, N = 3SE +/- 1.89, N = 3SE +/- 4.67, N = 3SE +/- 22.14, N = 3SE +/- 19.64, N = 3SE +/- 10.58, N = 3SE +/- 22.89, N = 3SE +/- 23.93, N = 3SE +/- 89.65, N = 3SE +/- 12.67, N = 3SE +/- 12.15, N = 3SE +/- 69.94, N = 3SE +/- 23.24, N = 3SE +/- 52.64, N = 3SE +/- 9.59, N = 3SE +/- 4.01, N = 3SE +/- 6.44, N = 3SE +/- 236.37, N = 3SE +/- 67.98, N = 3SE +/- 12.20, N = 3SE +/- 41.82, N = 3SE +/- 173.63, N = 316465.1218906.3314538.4216738.5922827.2045325.2128310.8255966.7562243.0726495.6351865.1261028.6734497.5167296.5773195.7412751.229537.9718459.4818809.5533544.0820486.9736840.381. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

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: BlowfishEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P90K180K270K360K450KSE +/- 83.93, N = 3SE +/- 215.41, N = 3SE +/- 29.80, N = 3SE +/- 178.00, N = 3SE +/- 11.89, N = 3SE +/- 4421.67, N = 15SE +/- 18.68, N = 3SE +/- 2993.24, N = 15SE +/- 12114.05, N = 13SE +/- 18.85, N = 3SE +/- 880.66, N = 3SE +/- 2077.77, N = 3SE +/- 493.13, N = 3SE +/- 4540.15, N = 15SE +/- 4832.61, N = 5SE +/- 455.51, N = 14SE +/- 6.67, N = 3SE +/- 405.72, N = 3SE +/- 10.67, N = 3SE +/- 856.25, N = 3SE +/- 142.00, N = 3SE +/- 2641.53, N = 15105587114112942661121151380322437231668873115213520911660803238553773182046774126024351937767057011112458939131781121042321895361. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Wide Vector MathEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2M4M6M8M10MSE +/- 116.18, N = 3SE +/- 132.65, N = 3SE +/- 90.55, N = 3SE +/- 2855.52, N = 3SE +/- 1186.05, N = 3SE +/- 30.20, N = 3SE +/- 41.80, N = 3SE +/- 4084.81, N = 3SE +/- 31722.54, N = 3SE +/- 30.71, N = 3SE +/- 2431.22, N = 3SE +/- 11790.59, N = 3SE +/- 495.80, N = 3SE +/- 1194.76, N = 3SE +/- 7486.66, N = 3SE +/- 1151.93, N = 3SE +/- 394.58, N = 3SE +/- 962.64, N = 3SE +/- 3353.99, N = 3SE +/- 9436.99, N = 3SE +/- 3060.01, N = 3SE +/- 69046.18, N = 31667888.592018031.721406939.121602887.522250265.274381848.832597630.055107325.666004812.792597371.595100138.715929288.233437154.966653978.007475976.331466592.911136667.112227716.194423715.197733151.594735239.998548723.361. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

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: bcryptEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P90K180K270K360K450KSE +/- 21.08, N = 3SE +/- 12.67, N = 3SE +/- 48.67, N = 3SE +/- 353.20, N = 3SE +/- 33.83, N = 3SE +/- 3452.16, N = 3SE +/- 69.15, N = 3SE +/- 4708.84, N = 15SE +/- 1544.85, N = 3SE +/- 0.67, N = 3SE +/- 4919.29, N = 15SE +/- 3106.56, N = 3SE +/- 152.02, N = 3SE +/- 1504.30, N = 3SE +/- 5319.74, N = 15SE +/- 56.89, N = 3SE +/- 8.19, N = 3SE +/- 89.08, N = 3SE +/- 30.49, N = 3SE +/- 784.76, N = 3SE +/- 272.32, N = 3SE +/- 1964.45, N = 6105831114572942291115651380352471151668503082073788121661193090913739062048284219774284947821456989113216938771781081040671963981. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Pabellon Barcelona - Compute: CPU-OnlyEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.27, N = 3SE +/- 0.16, N = 3SE +/- 0.11, N = 3SE +/- 0.23, N = 3SE +/- 0.22, N = 3SE +/- 0.06, N = 3SE +/- 0.29, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 3SE +/- 0.06, N = 3SE +/- 0.27, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.20, N = 3SE +/- 0.34, N = 3SE +/- 0.17, N = 3SE +/- 0.16, N = 3SE +/- 0.28, N = 3SE +/- 0.34, N = 380.7168.2193.3384.3157.5431.1448.7725.7022.7349.8426.6023.3339.5321.7519.92115.90148.8676.3188.5147.1681.1643.30

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.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2K4K6K8K10KSE +/- 0.11, N = 3SE +/- 0.70, N = 3SE +/- 0.28, N = 3SE +/- 9.73, N = 3SE +/- 34.21, N = 3SE +/- 58.12, N = 15SE +/- 13.67, N = 3SE +/- 61.26, N = 4SE +/- 60.74, N = 4SE +/- 0.19, N = 3SE +/- 70.92, N = 4SE +/- 58.90, N = 4SE +/- 1.57, N = 3SE +/- 36.84, N = 4SE +/- 3.16, N = 4SE +/- 18.43, N = 3SE +/- 9.43, N = 3SE +/- 20.22, N = 3SE +/- 11.04, N = 3SE +/- 37.52, N = 3SE +/- 6.66, N = 3SE +/- 50.37, N = 152806.353167.602487.442976.003883.587369.804400.508300.429152.764251.207890.088928.995892.347870.837970.171556.281243.422386.402146.293544.572682.324875.551. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P80160240320400SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.39, N = 3SE +/- 1.37, N = 3SE +/- 2.32, N = 15SE +/- 0.55, N = 3SE +/- 2.45, N = 4SE +/- 2.43, N = 4SE +/- 0.01, N = 3SE +/- 2.84, N = 4SE +/- 2.36, N = 4SE +/- 0.06, N = 3SE +/- 1.47, N = 4SE +/- 0.13, N = 4SE +/- 0.74, N = 3SE +/- 0.38, N = 3SE +/- 0.81, N = 3SE +/- 0.44, N = 3SE +/- 1.50, N = 3SE +/- 0.27, N = 3SE +/- 2.01, N = 15112.25126.7099.50119.04155.34294.79176.02332.02366.11170.05315.60357.16235.69314.83318.8162.2549.7495.4685.85141.78107.29195.021. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Barbershop - Compute: CPU-OnlyEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P100200300400500SE +/- 0.13, N = 3SE +/- 0.45, N = 3SE +/- 0.24, N = 3SE +/- 0.19, N = 3SE +/- 0.21, N = 3SE +/- 0.51, N = 3SE +/- 0.19, N = 3SE +/- 0.06, N = 3SE +/- 0.22, N = 3SE +/- 0.13, N = 3SE +/- 0.23, N = 3SE +/- 0.10, N = 3SE +/- 0.37, N = 3SE +/- 0.05, N = 3SE +/- 0.15, N = 3SE +/- 0.07, N = 3SE +/- 0.35, N = 3SE +/- 0.33, N = 3SE +/- 0.24, N = 3SE +/- 0.21, N = 3SE +/- 0.07, N = 3SE +/- 4.74, N = 12236.61203.64267.69241.01170.5696.44142.3882.4872.20142.5477.9371.09118.4468.9162.68349.23459.67239.18273.44146.55244.98153.00

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 SecondEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2M4M6M8M10MSE +/- 3641.68, N = 3SE +/- 15281.40, N = 3SE +/- 5062.68, N = 3SE +/- 9763.09, N = 3SE +/- 19980.87, N = 3SE +/- 3921.61, N = 3SE +/- 9714.51, N = 3SE +/- 3868.89, N = 3SE +/- 17358.34, N = 3SE +/- 5924.03, N = 3SE +/- 10397.88, N = 3SE +/- 21137.20, N = 3SE +/- 25359.52, N = 3SE +/- 50863.65, N = 3SE +/- 20407.34, N = 3SE +/- 6091.74, N = 3SE +/- 12498.07, N = 3SE +/- 19480.38, N = 3SE +/- 1523.22, N = 3SE +/- 8498.29, N = 3SE +/- 3253.53, N = 3SE +/- 8766.38, N = 32415599.082593724.282152972.962484026.793056532.145801339.813884296.778035017.428770195.343703841.727647272.558618342.534543138.128228175.708229554.421569462.321200098.072281894.062233945.614492941.732410201.885001821.751. (CC) gcc options: -O2 -lrt" -lrt

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/ and https://github.com/OpenRadioss/ModelExchange/tree/main/Examples. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Chrysler Neon 1MEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P70140210280350SE +/- 0.24, N = 3SE +/- 0.57, N = 3SE +/- 0.25, N = 3SE +/- 0.18, N = 3SE +/- 1.64, N = 12SE +/- 1.06, N = 12SE +/- 1.75, N = 12SE +/- 0.54, N = 11SE +/- 0.67, N = 15SE +/- 0.44, N = 3SE +/- 0.18, N = 3SE +/- 0.04, N = 3SE +/- 2.82, N = 9SE +/- 1.48, N = 11SE +/- 0.17, N = 3SE +/- 0.21, N = 3SE +/- 0.28, N = 3SE +/- 0.29, N = 3SE +/- 0.19, N = 3SE +/- 0.18, N = 3SE +/- 0.07, N = 3SE +/- 0.24, N = 3295.49296.41301.17296.46204.8785.72176.2871.4867.6686.9853.5346.95212.0890.0573.77257.84342.74182.96205.86114.30157.1784.09

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 0.13Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P11K22K33K44K55KSE +/- 155.56, N = 3SE +/- 344.15, N = 3SE +/- 129.97, N = 3SE +/- 346.70, N = 3SE +/- 22.88, N = 3SE +/- 23.33, N = 3SE +/- 14.80, N = 3SE +/- 19.06, N = 3SE +/- 10.68, N = 3SE +/- 30.40, N = 3SE +/- 63.54, N = 3SE +/- 6.74, N = 3SE +/- 11.92, N = 3SE +/- 6.43, N = 3SE +/- 7.54, N = 3SE +/- 130.37, N = 3SE +/- 313.52, N = 3SE +/- 29.55, N = 3SE +/- 162.18, N = 3SE +/- 38.25, N = 3SE +/- 8.69, N = 3SE +/- 159.27, N = 3370503272940962356552208312290182651027587421819199548624147528205739747446529383011931047140432286012079

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P11K22K33K44K55KSE +/- 152.18, N = 3SE +/- 184.20, N = 3SE +/- 220.42, N = 3SE +/- 138.54, N = 3SE +/- 18.82, N = 3SE +/- 9.61, N = 3SE +/- 55.52, N = 3SE +/- 26.46, N = 3SE +/- 7.22, N = 3SE +/- 29.57, N = 3SE +/- 41.29, N = 3SE +/- 20.98, N = 3SE +/- 21.08, N = 3SE +/- 11.98, N = 3SE +/- 10.40, N = 3SE +/- 45.13, N = 3SE +/- 89.07, N = 3SE +/- 60.34, N = 3SE +/- 123.98, N = 3SE +/- 10.17, N = 3SE +/- 59.69, N = 3SE +/- 144.98, N = 33711533089411333594222356124811843510431885118269100498692149488329751748287535163037431234142322298012215

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P9K18K27K36K45KSE +/- 3.53, N = 3SE +/- 5.87, N = 3SE +/- 2.53, N = 3SE +/- 5.77, N = 3SE +/- 2.03, N = 3SE +/- 11.03, N = 3SE +/- 10.80, N = 3SE +/- 4.25, N = 3SE +/- 60.83, N = 3SE +/- 3.00, N = 3SE +/- 0.83, N = 3SE +/- 37.41, N = 3SE +/- 10.34, N = 3SE +/- 17.53, N = 3SE +/- 18.00, N = 3SE +/- 10.53, N = 3SE +/- 7.78, N = 3SE +/- 8.11, N = 3SE +/- 0.87, N = 3SE +/- 12.77, N = 3SE +/- 4.27, N = 3SE +/- 28.91, N = 39775.4411615.148305.178850.9013514.5226807.7916149.9932601.9136200.3714961.4931334.5836106.6020080.5439496.6242407.408143.605978.0613525.459340.5521058.4012358.7424813.701. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P10002000300040005000SE +/- 1.72, N = 3SE +/- 6.20, N = 3SE +/- 2.12, N = 3SE +/- 2.07, N = 3SE +/- 2.16, N = 3SE +/- 1.97, N = 3SE +/- 3.64, N = 3SE +/- 1.39, N = 3SE +/- 4.57, N = 3SE +/- 0.90, N = 3SE +/- 1.84, N = 3SE +/- 5.82, N = 3SE +/- 36.44, N = 13SE +/- 1.77, N = 3SE +/- 4.71, N = 3SE +/- 0.43, N = 3SE +/- 0.62, N = 3SE +/- 1.54, N = 3SE +/- 0.56, N = 3SE +/- 2.83, N = 3SE +/- 1.77, N = 3SE +/- 3.34, N = 31046.381168.48938.50971.681550.492853.561879.953488.923994.651915.163436.394022.341708.774202.084577.35767.66646.031194.861098.311909.271252.232388.201. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 0.13Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P13K26K39K52K65KSE +/- 135.93, N = 3SE +/- 209.35, N = 3SE +/- 178.39, N = 3SE +/- 190.84, N = 3SE +/- 104.80, N = 3SE +/- 21.88, N = 3SE +/- 32.27, N = 3SE +/- 21.18, N = 3SE +/- 19.67, N = 3SE +/- 27.81, N = 3SE +/- 34.96, N = 3SE +/- 57.71, N = 3SE +/- 23.95, N = 3SE +/- 24.01, N = 3SE +/- 16.62, N = 3SE +/- 37.04, N = 3SE +/- 40.48, N = 3SE +/- 44.47, N = 3SE +/- 49.64, N = 3SE +/- 67.91, N = 3SE +/- 17.91, N = 3SE +/- 18.94, N = 3426643793647070412253123614599215501213510364214081168010240175519762883755097619823459035948167373141514106

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: CPU StressEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120K240K360K480K600KSE +/- 83.94, N = 3SE +/- 22.33, N = 3SE +/- 30.20, N = 3SE +/- 279.91, N = 3SE +/- 284.05, N = 3SE +/- 12266.01, N = 15SE +/- 350.09, N = 3SE +/- 3681.41, N = 3SE +/- 896.12, N = 3SE +/- 69.19, N = 3SE +/- 8871.14, N = 15SE +/- 1441.70, N = 3SE +/- 173.58, N = 3SE +/- 11493.27, N = 12SE +/- 526.04, N = 3SE +/- 140.44, N = 3SE +/- 93.31, N = 3SE +/- 3794.61, N = 12SE +/- 185.13, N = 3SE +/- 296.85, N = 3SE +/- 150.21, N = 3SE +/- 416.60, N = 3132960.69145093.05118644.62135689.25116427.04271343.96220744.88326382.18474925.60211280.23366232.61463915.94278579.23495158.04557160.74109916.2780158.11153923.09146387.61260500.59156861.89280684.861. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Floating PointEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P17K34K51K68K85KSE +/- 29.45, N = 3SE +/- 2.23, N = 3SE +/- 29.39, N = 3SE +/- 44.16, N = 3SE +/- 2.37, N = 3SE +/- 29.27, N = 3SE +/- 41.92, N = 3SE +/- 247.68, N = 3SE +/- 64.92, N = 3SE +/- 26.43, N = 3SE +/- 73.54, N = 3SE +/- 67.44, N = 3SE +/- 56.54, N = 3SE +/- 129.04, N = 3SE +/- 10.84, N = 3SE +/- 3.46, N = 3SE +/- 6.80, N = 3SE +/- 1.61, N = 3SE +/- 13.87, N = 3SE +/- 68.25, N = 3SE +/- 23.45, N = 3SE +/- 49.95, N = 318641.7519966.2716586.4319466.4624168.3647698.1430824.9760101.6967867.7229220.7857545.5366261.1839031.6475275.5977279.8411247.8811125.8522002.8222454.7039566.7023555.9243220.201. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: BMW27 - Compute: CPU-OnlyEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 5SE +/- 0.07, N = 4SE +/- 0.04, N = 5SE +/- 0.02, N = 6SE +/- 0.04, N = 3SE +/- 0.04, N = 5SE +/- 0.04, N = 5SE +/- 0.03, N = 4SE +/- 0.04, N = 6SE +/- 0.03, N = 6SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.12, N = 3SE +/- 0.10, N = 4SE +/- 0.05, N = 3SE +/- 0.06, N = 426.0121.7630.0926.5118.7110.2815.898.657.6216.288.787.8712.997.326.7134.4945.8423.6225.9113.9623.9612.70

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 0.13Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20K40K60K80K100KSE +/- 112.90, N = 3SE +/- 152.73, N = 3SE +/- 209.22, N = 3SE +/- 197.95, N = 3SE +/- 30.95, N = 3SE +/- 711.60, N = 15SE +/- 177.18, N = 3SE +/- 72.02, N = 3SE +/- 26.83, N = 3SE +/- 48.77, N = 3SE +/- 61.66, N = 3SE +/- 24.58, N = 3SE +/- 6.36, N = 3SE +/- 42.71, N = 3SE +/- 44.41, N = 3SE +/- 55.37, N = 3SE +/- 165.65, N = 3SE +/- 56.55, N = 3SE +/- 15.94, N = 3SE +/- 247.53, N = 3SE +/- 62.20, N = 3SE +/- 36.67, N = 3676715936075507652164969926505419422058117500421831990317213360661651214814855851002675458657203336455007023845

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20K40K60K80K100KSE +/- 358.36, N = 3SE +/- 183.64, N = 3SE +/- 235.50, N = 3SE +/- 133.70, N = 3SE +/- 35.48, N = 3SE +/- 642.30, N = 15SE +/- 223.77, N = 3SE +/- 23.88, N = 3SE +/- 58.18, N = 3SE +/- 182.72, N = 3SE +/- 23.13, N = 3SE +/- 10.02, N = 3SE +/- 158.22, N = 3SE +/- 28.26, N = 3SE +/- 56.37, N = 3SE +/- 200.14, N = 3SE +/- 662.96, N = 3SE +/- 115.90, N = 3SE +/- 166.38, N = 3SE +/- 321.54, N = 3SE +/- 42.45, N = 3SE +/- 80.34, N = 3677516010475559660394989528937425202073817787419161997117396361361670215084876231012615498857985340295024224038

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 102.71, N = 3SE +/- 208.73, N = 3SE +/- 284.23, N = 3SE +/- 267.91, N = 3SE +/- 94.72, N = 3SE +/- 273.26, N = 3SE +/- 195.23, N = 3SE +/- 767.79, N = 15SE +/- 41.68, N = 3SE +/- 38.15, N = 3SE +/- 73.51, N = 3SE +/- 48.08, N = 3SE +/- 285.33, N = 3SE +/- 37.82, N = 3SE +/- 34.67, N = 3SE +/- 72.73, N = 3SE +/- 222.59, N = 3SE +/- 255.41, N = 3SE +/- 181.03, N = 3SE +/- 213.87, N = 3SE +/- 48.56, N = 3SE +/- 182.14, N = 37908369250877897629757960345734850527766207504845023284203244154619542176621012351171486369867421390555847733070

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3K6K9K12K15KSE +/- 1.37, N = 3SE +/- 5.60, N = 3SE +/- 0.53, N = 3SE +/- 4.19, N = 3SE +/- 1.69, N = 3SE +/- 2.71, N = 3SE +/- 2.20, N = 3SE +/- 3.32, N = 3SE +/- 6.29, N = 3SE +/- 2.25, N = 3SE +/- 1.55, N = 3SE +/- 7.78, N = 3SE +/- 69.80, N = 13SE +/- 1.37, N = 3SE +/- 17.79, N = 3SE +/- 2.47, N = 3SE +/- 1.53, N = 3SE +/- 6.18, N = 3SE +/- 2.94, N = 3SE +/- 9.83, N = 3SE +/- 0.63, N = 3SE +/- 4.96, N = 33123.393503.912813.702914.864691.888978.455665.0810933.9411978.805425.6510367.4811860.855762.2713075.2713902.662759.762125.064427.003231.665793.794694.228918.861. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 RateEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.13, N = 5SE +/- 0.12, N = 5SE +/- 0.06, N = 5SE +/- 0.25, N = 5SE +/- 0.28, N = 6SE +/- 0.14, N = 8SE +/- 0.18, N = 7SE +/- 0.19, N = 9SE +/- 0.19, N = 9SE +/- 0.12, N = 7SE +/- 0.06, N = 9SE +/- 0.28, N = 9SE +/- 0.13, N = 7SE +/- 0.14, N = 10SE +/- 0.12, N = 10SE +/- 0.16, N = 15SE +/- 0.01, N = 3SE +/- 0.09, N = 5SE +/- 0.13, N = 5SE +/- 0.31, N = 7SE +/- 0.04, N = 6SE +/- 0.06, N = 822.8223.0521.1522.8630.1754.7439.2869.6779.8240.3269.4078.9843.6886.3388.1319.3913.6224.9723.2645.7929.1458.641. (CC) gcc options: -O3 -march=native -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 0.13Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P6001200180024003000SE +/- 0.58, N = 3SE +/- 1.33, N = 3SE +/- 1.15, N = 3SE +/- 4.16, N = 3SE +/- 3.51, N = 3SE +/- 1.15, N = 3SE +/- 0.00, N = 3SE +/- 1.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 1.53, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 1.33, N = 3SE +/- 4.06, N = 3SE +/- 18.15, N = 3SE +/- 1.33, N = 3SE +/- 0.88, N = 3SE +/- 1.20, N = 3SE +/- 1.20, N = 3SE +/- 1.20, N = 3190516572145183314017731142644547113262354092451746224172983152816458861427748

libxsmm

Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 128EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P15003000450060007500SE +/- 0.40, N = 3SE +/- 0.80, N = 3SE +/- 1.56, N = 3SE +/- 1.22, N = 3SE +/- 3.26, N = 3SE +/- 12.97, N = 3SE +/- 1.99, N = 3SE +/- 107.43, N = 9SE +/- 10.63, N = 3SE +/- 10.12, N = 3SE +/- 318.58, N = 9SE +/- 65.29, N = 3SE +/- 8.00, N = 3SE +/- 76.63, N = 6SE +/- 147.77, N = 9SE +/- 6.03, N = 3SE +/- 0.07, N = 3SE +/- 16.14, N = 3SE +/- 2.52, N = 3SE +/- 25.10, N = 9SE +/- 32.41, N = 9SE +/- 83.06, N = 61304.91321.71244.91305.52400.44031.32578.64275.84749.73102.45420.67125.32702.53982.04190.81431.81110.72006.11764.62948.01873.43209.01. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Matrix MathEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P200K400K600K800K1000KSE +/- 629.51, N = 3SE +/- 23.05, N = 3SE +/- 412.52, N = 3SE +/- 441.17, N = 3SE +/- 6782.96, N = 12SE +/- 61.00, N = 3SE +/- 508.29, N = 3SE +/- 11090.95, N = 12SE +/- 2638.97, N = 3SE +/- 944.14, N = 3SE +/- 1709.79, N = 3SE +/- 5881.56, N = 3SE +/- 236.23, N = 3SE +/- 5101.58, N = 3SE +/- 1186.29, N = 3SE +/- 123.69, N = 3SE +/- 804.31, N = 3SE +/- 3842.87, N = 3SE +/- 3004.11, N = 6SE +/- 4330.71, N = 15SE +/- 4887.40, N = 15SE +/- 1338.72, N = 3266664.84311884.13233716.23259675.50346518.02688593.49421053.02816181.03965326.58417406.39817035.07939452.37552067.041062616.621164762.86258693.38181695.08353381.32304345.30535739.58301894.53536146.581. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

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 0.13Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P8001600240032004000SE +/- 1.33, N = 3SE +/- 2.03, N = 3SE +/- 2.03, N = 3SE +/- 2.31, N = 3SE +/- 0.58, N = 3SE +/- 1.67, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 1.15, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 0.67, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 5.86, N = 3SE +/- 20.48, N = 3SE +/- 2.73, N = 3SE +/- 0.58, N = 3SE +/- 1.67, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 322611973254621851671916134876564813357326391097616554290635211803196510591691887

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.13Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P6001200180024003000SE +/- 1.20, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 3.38, N = 3SE +/- 0.88, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.67, N = 3SE +/- 1.76, N = 3SE +/- 1.53, N = 3SE +/- 1.76, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 2.08, N = 3SE +/- 15.70, N = 3SE +/- 1.86, N = 3SE +/- 1.00, N = 3SE +/- 1.67, N = 3SE +/- 0.33, N = 3SE +/- 1.00, N = 3192216782163185714127851149653554114262854893452647024672982154016629061430751

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 2023.2.devModel: Person Detection FP32 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300SE +/- 0.17, N = 3SE +/- 0.60, N = 3SE +/- 0.19, N = 3SE +/- 0.19, N = 3SE +/- 0.95, N = 15SE +/- 0.18, N = 3SE +/- 1.56, N = 14SE +/- 0.61, N = 3SE +/- 0.19, N = 3SE +/- 0.78, N = 11SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 3.38, N = 13SE +/- 0.59, N = 15SE +/- 0.75, N = 14SE +/- 0.18, N = 3SE +/- 0.47, N = 3SE +/- 0.28, N = 3SE +/- 0.12, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3144.93146.47162.65148.15105.9147.93153.1160.2350.78106.9048.4043.41257.8285.2083.88133.91238.41114.61103.3346.8540.8842.551. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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.CEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P130K260K390K520K650KSE +/- 580.25, N = 4SE +/- 325.27, N = 4SE +/- 346.68, N = 4SE +/- 343.36, N = 4SE +/- 1817.56, N = 5SE +/- 3935.08, N = 6SE +/- 686.23, N = 5SE +/- 2062.83, N = 6SE +/- 2260.16, N = 6SE +/- 1937.17, N = 6SE +/- 1082.71, N = 6SE +/- 1962.23, N = 6SE +/- 1940.05, N = 15SE +/- 6881.36, N = 15SE +/- 6909.67, N = 15SE +/- 170.50, N = 3SE +/- 8.86, N = 3SE +/- 195.01, N = 4SE +/- 157.58, N = 4SE +/- 590.76, N = 5SE +/- 424.75, N = 5SE +/- 3797.43, N = 5137060.08140067.08132191.02139476.20253193.84476679.49271216.04497487.00524075.11333543.81516601.66569289.36279880.91596937.78616796.98115269.9998350.98185826.45136342.91263384.92233511.62462410.561. (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 4.1.5

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Fishy Cat - Compute: CPU-OnlyEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1326395265SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 4SE +/- 0.08, N = 3SE +/- 0.08, N = 4SE +/- 0.07, N = 5SE +/- 0.07, N = 3SE +/- 0.10, N = 4SE +/- 0.06, N = 5SE +/- 0.06, N = 3SE +/- 0.03, N = 5SE +/- 0.07, N = 5SE +/- 0.09, N = 3SE +/- 0.01, N = 3SE +/- 0.26, N = 3SE +/- 0.32, N = 3SE +/- 0.27, N = 3SE +/- 0.23, N = 3SE +/- 0.04, N = 333.1428.2738.0834.8023.9913.5620.3211.5310.5620.6911.6310.5916.8410.189.4245.2358.6630.8635.6519.7032.7017.57

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 2023.2.devModel: Person Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.20, N = 3SE +/- 0.33, N = 3SE +/- 0.20, N = 3SE +/- 0.28, N = 3SE +/- 0.15, N = 3SE +/- 0.02, N = 3SE +/- 3.35, N = 15SE +/- 0.41, N = 15SE +/- 0.35, N = 14SE +/- 0.02, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 6.66, N = 15SE +/- 1.50, N = 15SE +/- 1.70, N = 15SE +/- 0.65, N = 3SE +/- 0.32, N = 3SE +/- 0.24, N = 3SE +/- 0.18, N = 3SE +/- 0.09, N = 3SE +/- 0.15, N = 3SE +/- 0.06, N = 3145.25147.42162.25148.1494.0447.89150.3160.4251.40100.5148.6143.37251.0584.3182.03133.20238.89113.75103.4646.9340.9942.541. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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: BT.CEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P150K300K450K600K750KSE +/- 74.17, N = 3SE +/- 130.35, N = 3SE +/- 145.57, N = 3SE +/- 336.84, N = 3SE +/- 560.21, N = 4SE +/- 3427.71, N = 5SE +/- 1485.73, N = 4SE +/- 2471.37, N = 5SE +/- 3842.71, N = 5SE +/- 1185.12, N = 5SE +/- 5519.23, N = 5SE +/- 5727.88, N = 15SE +/- 174.81, N = 4SE +/- 6256.93, N = 15SE +/- 4878.86, N = 6SE +/- 224.88, N = 3SE +/- 146.25, N = 3SE +/- 239.40, N = 4SE +/- 44.76, N = 3SE +/- 871.29, N = 5SE +/- 161.64, N = 4SE +/- 2706.26, N = 5137005.01137649.22133857.43137749.25233622.44452577.13261864.11509783.05551536.08312135.73625624.70708202.58296367.04507704.84514956.56134960.24115701.63192791.12180757.43336631.69236490.86437263.741. (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 4.1.5

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 Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Wownero - Hash Count: 1MEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 111.40, N = 3SE +/- 157.30, N = 3SE +/- 21.56, N = 3SE +/- 89.34, N = 3SE +/- 18.19, N = 3SE +/- 114.08, N = 4SE +/- 60.06, N = 4SE +/- 246.37, N = 5SE +/- 312.48, N = 5SE +/- 35.19, N = 4SE +/- 202.10, N = 5SE +/- 239.10, N = 5SE +/- 604.78, N = 4SE +/- 589.33, N = 5SE +/- 763.41, N = 5SE +/- 157.77, N = 3SE +/- 79.10, N = 3SE +/- 104.40, N = 3SE +/- 13.06, N = 3SE +/- 87.84, N = 3SE +/- 13.47, N = 3SE +/- 491.32, N = 439364.042491.135279.640241.860032.498348.175127.4125526.0136083.273567.3119142.6131097.777607.2138322.7144416.427482.923745.641014.435870.260698.342867.674961.11. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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 digitEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 3SE +/- 0.20, N = 3SE +/- 0.12, N = 3SE +/- 0.44, N = 4SE +/- 0.32, N = 3SE +/- 0.28, N = 3SE +/- 0.03, N = 3SE +/- 0.27, N = 3SE +/- 0.08, N = 3SE +/- 0.13, N = 3SE +/- 0.21, N = 3SE +/- 0.31, N = 5SE +/- 0.36, N = 3SE +/- 0.11, N = 3SE +/- 0.11, N = 3SE +/- 0.34, N = 3SE +/- 0.43, N = 3SE +/- 0.64, N = 3SE +/- 0.09, N = 3SE +/- 0.28, N = 394.4278.94113.14103.6971.2738.1961.5331.6628.8764.3532.6729.3348.9528.0826.70143.63161.8984.5286.7650.4884.9549.321. (CC) gcc options: -O2 -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.DEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3K6K9K12K15KSE +/- 6.53, N = 4SE +/- 17.45, N = 4SE +/- 10.84, N = 4SE +/- 14.78, N = 4SE +/- 14.93, N = 5SE +/- 71.24, N = 6SE +/- 13.89, N = 5SE +/- 85.70, N = 8SE +/- 90.82, N = 6SE +/- 65.80, N = 5SE +/- 79.11, N = 6SE +/- 117.09, N = 15SE +/- 18.26, N = 5SE +/- 76.17, N = 13SE +/- 125.43, N = 4SE +/- 8.92, N = 3SE +/- 7.21, N = 3SE +/- 7.24, N = 4SE +/- 6.63, N = 4SE +/- 49.33, N = 4SE +/- 13.00, N = 4SE +/- 196.02, N = 152757.072798.202707.252774.415184.369783.655195.2410049.349969.476138.5411652.1612345.625773.2511222.7011885.212060.062333.983338.682885.713954.753974.937849.171. (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 4.1.5

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Fused Multiply-AddEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120M240M360M480M600MSE +/- 223402.69, N = 3SE +/- 140896.40, N = 3SE +/- 247735.79, N = 3SE +/- 240460.19, N = 3SE +/- 1366328.72, N = 3SE +/- 263500.46, N = 3SE +/- 301328.09, N = 3SE +/- 561393.24, N = 3SE +/- 681486.02, N = 3SE +/- 140980.34, N = 3SE +/- 2347159.80, N = 3SE +/- 389118.11, N = 3SE +/- 2076871.25, N = 3SE +/- 399777.22, N = 3SE +/- 1475361.84, N = 3SE +/- 581877.08, N = 3SE +/- 52085.09, N = 3SE +/- 86473.30, N = 3SE +/- 222901.64, N = 3SE +/- 328083.91, N = 3SE +/- 131220.52, N = 3SE +/- 245288.30, N = 3124404567.22149843902.81106714867.67118913171.76167642337.90333205139.82202941117.45399961883.83453551958.37202387450.78393382814.53442132595.98256387572.72497503915.39554322523.21112663200.8392550459.33181639452.04145542113.66254737912.80159324513.29292027758.581. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

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 DirectionEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9754EPYC 9754 2PXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.03409461, N = 3SE +/- 0.05891794, N = 3SE +/- 0.06837566, N = 3SE +/- 0.03648952, N = 3SE +/- 0.03040517, N = 5SE +/- 0.02698811, N = 6SE +/- 0.07670149, N = 5SE +/- 0.03748638, N = 5SE +/- 0.00774286, N = 5SE +/- 0.03937117, N = 5SE +/- 0.03515823, N = 5SE +/- 0.04030589, N = 5SE +/- 0.01017307, N = 4SE +/- 0.03645493, N = 3SE +/- 0.02480228, N = 3SE +/- 0.00347933, N = 4SE +/- 0.04713862, N = 4SE +/- 0.02537673, N = 5SE +/- 0.03576289, N = 4SE +/- 0.05600796, N = 520.9124565020.5405852020.8184172020.787467909.745516214.876868178.452129364.293908984.193066027.308144194.023657859.028061684.4138232518.4491412024.0121949011.0611921012.798342006.1712420510.181469405.246908001. (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

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 PSKEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P300K600K900K1200K1500KSE +/- 283.35, N = 3SE +/- 240.76, N = 3SE +/- 363.67, N = 3SE +/- 79.00, N = 3SE +/- 23.00, N = 3SE +/- 22348.77, N = 12SE +/- 64.33, N = 3SE +/- 15108.39, N = 15SE +/- 1763.83, N = 3SE +/- 68.67, N = 3SE +/- 27844.49, N = 12SE +/- 15169.41, N = 3SE +/- 2344.03, N = 3SE +/- 33042.67, N = 15SE +/- 38452.07, N = 15SE +/- 259.89, N = 3SE +/- 129.95, N = 3SE +/- 3175.26, N = 3SE +/- 184.59, N = 3SE +/- 8456.62, N = 3SE +/- 78.84, N = 3SE +/- 10079.77, N = 124022294182893407413596395127739500026493451222333133733361160111371681302333779799148153314970003759172520764881103991417661224287107973591. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

rays1bench

This is a test of rays1bench, a simple path-tracer / ray-tracing that supports SSE and AVX instructions, multi-threading, and other features. This test profile is measuring the performance of the "large scene" in rays1bench. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmrays/s, More Is Betterrays1bench 2020-01-09Large SceneEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2004006008001000SE +/- 0.37, N = 8SE +/- 0.44, N = 8SE +/- 0.19, N = 7SE +/- 0.40, N = 8SE +/- 0.56, N = 9SE +/- 2.01, N = 10SE +/- 0.61, N = 9SE +/- 3.60, N = 10SE +/- 3.83, N = 10SE +/- 0.59, N = 9SE +/- 2.05, N = 10SE +/- 3.49, N = 10SE +/- 1.13, N = 9SE +/- 4.17, N = 10SE +/- 5.39, N = 10SE +/- 2.11, N = 15SE +/- 0.29, N = 7SE +/- 0.61, N = 8SE +/- 1.94, N = 8SE +/- 1.66, N = 9SE +/- 1.17, N = 9SE +/- 3.31, N = 10341.28406.06270.54306.52469.77803.65541.15893.47965.94519.38863.84952.84587.38998.921077.62238.53182.32333.97334.77611.85378.27660.98

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 2023.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.00, N = 3SE +/- 0.02, 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 +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.12, N = 13SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 310.139.0111.2610.866.723.558.344.373.998.734.614.0411.014.884.595.009.214.5018.3720.6313.6214.341. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 2.12Benchmark: gravity_spheres_volume/dim_512/ao/real_timeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1326395265SE +/- 0.00382, N = 3SE +/- 0.01769, N = 3SE +/- 0.02015, N = 3SE +/- 0.02542, N = 3SE +/- 0.03278, N = 3SE +/- 0.02296, N = 3SE +/- 0.01272, N = 3SE +/- 0.03171, N = 3SE +/- 0.03512, N = 3SE +/- 0.02462, N = 3SE +/- 0.06092, N = 3SE +/- 0.12462, N = 3SE +/- 0.01254, N = 3SE +/- 0.10903, N = 3SE +/- 0.14643, N = 3SE +/- 0.02876, N = 3SE +/- 0.02792, N = 3SE +/- 0.06239, N = 3SE +/- 0.10109, N = 3SE +/- 0.12438, N = 3SE +/- 0.07018, N = 3SE +/- 0.10393, N = 316.7063017.9160014.9512017.3083021.3127037.2318026.1498043.7516051.6200026.4152044.2117051.1259032.3187054.1369056.3885012.248209.8992618.9775019.3585036.5124022.8460042.90770

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.03816, N = 3SE +/- 0.01264, N = 3SE +/- 0.00526, N = 3SE +/- 0.02499, N = 3SE +/- 0.00538, N = 3SE +/- 0.09247, N = 3SE +/- 0.04369, N = 3SE +/- 0.10783, N = 3SE +/- 0.06365, N = 3SE +/- 0.03821, N = 3SE +/- 0.01787, N = 3SE +/- 0.10282, N = 3SE +/- 0.03723, N = 3SE +/- 0.07321, N = 3SE +/- 0.06789, N = 3SE +/- 0.04173, N = 3SE +/- 0.04904, N = 3SE +/- 0.02956, N = 3SE +/- 0.01150, N = 3SE +/- 0.04436, N = 3SE +/- 0.09943, N = 3SE +/- 0.00535, N = 316.1455017.4508014.4842016.7172020.7390036.0232025.5082042.5493049.8644025.7488042.9636049.5366031.0911052.6320055.0268012.059809.7141818.4412018.8086035.2898022.4134041.70540

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 5.79, N = 3SE +/- 95.79, N = 3SE +/- 47.17, N = 3SE +/- 154.48, N = 3SE +/- 60.19, N = 3SE +/- 582.33, N = 3SE +/- 116.54, N = 3SE +/- 1457.76, N = 3SE +/- 546.38, N = 3SE +/- 267.70, N = 3SE +/- 881.37, N = 3SE +/- 965.60, N = 3SE +/- 176.40, N = 3SE +/- 396.28, N = 3SE +/- 1112.59, N = 3SE +/- 48.82, N = 3SE +/- 42.88, N = 3SE +/- 26.31, N = 3SE +/- 89.40, N = 3SE +/- 111.26, N = 3SE +/- 160.00, N = 3SE +/- 422.60, N = 365468.5173571.8158065.1159330.5787092.81146208.6788979.21135374.63153305.2184074.22122698.67142203.3191245.02129173.90133758.9348898.5927152.2560574.4087274.95141769.5772499.68115238.241. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 2.12Benchmark: particle_volume/scivis/real_timeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1122334455SE +/- 0.04392, N = 3SE +/- 0.01355, N = 3SE +/- 0.00969, N = 3SE +/- 0.00471, N = 3SE +/- 0.00981, N = 3SE +/- 0.00459, N = 3SE +/- 0.02789, N = 3SE +/- 0.04077, N = 3SE +/- 0.06420, N = 3SE +/- 0.00888, N = 3SE +/- 0.00381, N = 3SE +/- 0.02730, N = 3SE +/- 0.01407, N = 3SE +/- 0.08605, N = 3SE +/- 0.02995, N = 3SE +/- 0.02245, N = 3SE +/- 0.05066, N = 3SE +/- 0.02405, N = 3SE +/- 0.01197, N = 3SE +/- 0.02473, N = 3SE +/- 0.02148, N = 3SE +/- 0.06879, N = 316.5898017.2554014.8370017.1527021.0432037.1378024.8894042.3845049.8003025.0401042.1003048.5468030.7263049.4949050.1541010.726408.9353716.3938019.9171034.6679023.5838041.73840

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 2023.2.devModel: Road Segmentation ADAS FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.05, N = 3SE +/- 0.14, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.65, N = 13SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 330.4827.2934.0032.8420.5811.2025.4713.7412.0025.0013.9511.9237.5315.2113.9718.2130.8516.7154.4862.7951.0753.561. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 2.12Benchmark: particle_volume/ao/real_timeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1122334455SE +/- 0.01554, N = 3SE +/- 0.04327, N = 3SE +/- 0.01269, N = 3SE +/- 0.01561, N = 3SE +/- 0.01579, N = 3SE +/- 0.00717, N = 3SE +/- 0.01516, N = 3SE +/- 0.04044, N = 3SE +/- 0.01186, N = 3SE +/- 0.00275, N = 3SE +/- 0.02135, N = 3SE +/- 0.01446, N = 3SE +/- 0.03832, N = 3SE +/- 0.05456, N = 3SE +/- 0.07856, N = 3SE +/- 0.07015, N = 3SE +/- 0.00309, N = 3SE +/- 0.03087, N = 3SE +/- 0.01302, N = 3SE +/- 0.06394, N = 3SE +/- 0.02969, N = 3SE +/- 0.06286, N = 316.6348017.2760014.8609017.2004021.1253037.2431024.9485042.3761049.8065025.0514042.1553048.5725030.8241049.6164050.2349010.676909.0189216.4445019.9509034.6433023.6065041.81770

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 2023.2.devModel: Face Detection Retail FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, 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 +/- 0.00, 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 +/- 0.00, N = 34.543.915.465.323.221.754.102.152.004.322.312.044.492.322.214.268.374.219.658.036.856.971. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, 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 +/- 0.00, N = 3SE +/- 0.01, 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 +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 311.229.2912.9711.937.898.019.839.798.9610.5610.539.0810.7110.829.978.218.518.213.512.522.372.441. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.11930.23860.35790.47720.5965SE +/- 0.00011, N = 3SE +/- 0.00054, N = 3SE +/- 0.00008, N = 3SE +/- 0.00019, N = 3SE +/- 0.00026, N = 4SE +/- 0.00076, N = 4SE +/- 0.00044, N = 4SE +/- 0.00098, N = 3SE +/- 0.00023, N = 3SE +/- 0.00070, N = 4SE +/- 0.00021, N = 3SE +/- 0.00075, N = 3SE +/- 0.00082, N = 4SE +/- 0.00027, N = 3SE +/- 0.00076, N = 3SE +/- 0.00089, N = 3SE +/- 0.00052, N = 3SE +/- 0.00015, N = 3SE +/- 0.00020, N = 3SE +/- 0.00232, N = 4SE +/- 0.00024, N = 4SE +/- 0.00022, N = 30.381740.331600.429950.385440.282270.186440.265360.129380.112480.265910.128890.112990.228360.107030.097580.378620.530420.281150.288120.246550.253640.18084

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: CrownEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.07, N = 4SE +/- 0.15, N = 5SE +/- 0.13, N = 4SE +/- 0.12, N = 5SE +/- 0.10, N = 5SE +/- 0.35, N = 7SE +/- 0.17, N = 6SE +/- 0.67, N = 8SE +/- 0.23, N = 8SE +/- 0.14, N = 6SE +/- 0.39, N = 8SE +/- 0.25, N = 8SE +/- 0.21, N = 6SE +/- 0.57, N = 8SE +/- 1.28, N = 8SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.15, N = 5SE +/- 0.13, N = 4SE +/- 0.36, N = 6SE +/- 0.10, N = 5SE +/- 0.22, N = 765.6673.8359.4568.6089.69153.87111.03179.83205.23117.46201.39226.87122.19206.61218.9946.0343.7882.6768.58123.3081.79151.77

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.15, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.14, N = 4SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 15SE +/- 0.10, N = 3SE +/- 0.15, N = 4SE +/- 0.09, N = 4SE +/- 0.06, N = 3SE +/- 0.19, N = 3SE +/- 0.20, N = 3SE +/- 0.53, N = 3SE +/- 0.12, N = 3SE +/- 0.39, N = 6SE +/- 0.25, N = 645.3838.1952.1246.1134.0120.5530.2418.2416.4430.3218.3216.6725.1516.4615.5359.6577.0844.5542.7126.6639.8924.781. (CXX) g++ options: -O2 -lOpenCL

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 4.0Preset: MediumEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120240360480600SE +/- 0.10, N = 7SE +/- 0.27, N = 7SE +/- 0.05, N = 7SE +/- 0.23, N = 7SE +/- 0.07, N = 8SE +/- 0.52, N = 8SE +/- 0.22, N = 8SE +/- 1.10, N = 8SE +/- 1.13, N = 8SE +/- 0.28, N = 8SE +/- 0.78, N = 8SE +/- 1.02, N = 8SE +/- 0.83, N = 7SE +/- 1.38, N = 7SE +/- 1.78, N = 7SE +/- 0.03, N = 6SE +/- 0.11, N = 5SE +/- 0.28, N = 6SE +/- 0.12, N = 7SE +/- 1.71, N = 7SE +/- 1.73, N = 7SE +/- 1.30, N = 8284.65313.92246.00265.06355.18514.77407.25464.46554.12398.98478.23518.79456.77475.99393.38151.59111.94211.28298.50487.78323.34536.081. (CXX) g++ options: -O3 -flto -pthread

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: 240 - Buffer Length: 256 - Filter Length: 512EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P600M1200M1800M2400M3000MSE +/- 173621.56, N = 3SE +/- 611918.66, N = 3SE +/- 519786.28, N = 3SE +/- 925388.81, N = 3SE +/- 723417.81, N = 3SE +/- 1401189.97, N = 3SE +/- 1044030.65, N = 3SE +/- 1287547.19, N = 3SE +/- 4475240.53, N = 3SE +/- 2195702.87, N = 3SE +/- 1300000.00, N = 3SE +/- 8038518.38, N = 3SE +/- 3752776.75, N = 3SE +/- 4792470.95, N = 3SE +/- 2110555.48, N = 3SE +/- 531632.71, N = 3SE +/- 595566.96, N = 3SE +/- 953939.20, N = 3SE +/- 1311911.24, N = 3SE +/- 4458076.32, N = 3SE +/- 366666.67, N = 3SE +/- 1440293.18, N = 3861733333103103333375158666793777666711760000002102200000140020000022591333332706666667133816666721664000002657966667171470000026367666672647333333738810000556820000106690000010137666671417566667118043333320640333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 ISPCEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P9001800270036004500SE +/- 1.45, N = 3SE +/- 0.58, N = 3SE +/- 0.67, N = 3SE +/- 0.88, N = 3SE +/- 13.17, N = 3SE +/- 1.76, N = 3SE +/- 1.45, N = 3SE +/- 3.48, N = 3SE +/- 1.20, N = 3SE +/- 0.88, N = 3SE +/- 3.93, N = 3SE +/- 2.40, N = 3SE +/- 2.85, N = 3SE +/- 7.69, N = 3SE +/- 7.42, N = 3SE +/- 1.20, N = 3SE +/- 1.67, N = 3SE +/- 0.88, N = 3SE +/- 2.00, N = 3SE +/- 12.33, N = 3SE +/- 0.33, N = 3SE +/- 1.86, N = 313261419119813581712280620993338385021453534396424904108430599490015561740268520123280

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsssp max_TEPS, More Is BetterGraph500 3.0Scale: 26EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P200M400M600M800M1000M336877000341837000318668000350727000467453000670675000499347000857107000948166000501222000825625000885000000502151000107479000010862800002286040003004110003908930004544830005215700005479150006539640001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: Asian DragonEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300SE +/- 0.08, N = 5SE +/- 0.09, N = 5SE +/- 0.05, N = 5SE +/- 0.08, N = 5SE +/- 0.04, N = 6SE +/- 0.71, N = 7SE +/- 0.11, N = 7SE +/- 0.14, N = 8SE +/- 0.59, N = 8SE +/- 0.05, N = 7SE +/- 0.42, N = 8SE +/- 0.54, N = 9SE +/- 0.28, N = 7SE +/- 1.14, N = 8SE +/- 0.63, N = 8SE +/- 0.02, N = 4SE +/- 0.07, N = 4SE +/- 0.21, N = 6SE +/- 0.06, N = 5SE +/- 0.13, N = 7SE +/- 0.19, N = 6SE +/- 0.59, N = 880.4186.9475.3183.82107.87176.21132.84212.93235.41143.10233.66262.20151.42249.84258.6559.2256.0699.7591.25152.49111.88199.15

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P90M180M270M360M450MSE +/- 836331.05, N = 3SE +/- 202848.22, N = 3SE +/- 147615.66, N = 3SE +/- 1857510.95, N = 3SE +/- 662770.52, N = 3SE +/- 1574113.89, N = 3SE +/- 612439.42, N = 3SE +/- 2412249.79, N = 3SE +/- 1756083.52, N = 3SE +/- 906405.74, N = 3SE +/- 4050750.45, N = 3SE +/- 1641510.40, N = 3SE +/- 2513657.44, N = 3SE +/- 863286.70, N = 3SE +/- 3836056.39, N = 3SE +/- 321156.89, N = 3SE +/- 486185.52, N = 3SE +/- 975757.17, N = 3SE +/- 819760.48, N = 3SE +/- 1253765.83, N = 3SE +/- 1389226.43, N = 3SE +/- 2525138.94, N = 615792063818045365514422935216260720621016705437049328325020902037233515540876460923392663136285945639850737929026050537575326838519765213480975888111984163860413130818594235032608154003409272155356

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 2023.2.devModel: Machine Translation EN To DE FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.17, N = 3SE +/- 0.34, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 3SE +/- 0.23, N = 3SE +/- 0.10, N = 3SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 3SE +/- 1.29, N = 12SE +/- 0.49, N = 3SE +/- 0.11, N = 3SE +/- 0.06, N = 3SE +/- 0.82, N = 3SE +/- 0.14, N = 3SE +/- 0.04, N = 3SE +/- 0.13, N = 3SE +/- 0.18, N = 3SE +/- 0.08, N = 397.0886.84107.98102.9168.6738.9685.2547.9345.4186.9748.8243.06119.1356.8849.4866.86127.9170.5164.6040.1428.1028.971. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 2023Implementation: MPI CPU - Input: water_GMX50_bareEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.005, N = 3SE +/- 0.011, N = 3SE +/- 0.008, N = 3SE +/- 0.025, N = 3SE +/- 0.003, N = 3SE +/- 0.175, N = 3SE +/- 0.008, N = 3SE +/- 0.007, N = 3SE +/- 0.018, N = 3SE +/- 0.346, N = 12SE +/- 0.049, N = 3SE +/- 0.050, N = 3SE +/- 0.019, N = 3SE +/- 0.057, N = 3SE +/- 0.006, N = 3SE +/- 0.002, N = 3SE +/- 0.024, N = 3SE +/- 0.019, N = 3SE +/- 0.003, N = 3SE +/- 0.078, N = 3SE +/- 0.010, N = 3SE +/- 0.045, N = 36.2606.7455.7186.0839.65116.94111.40818.96221.08211.54618.77521.10712.20219.81821.9025.8234.8439.0478.64515.22110.50418.3431. (CXX) g++ options: -O3

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.CEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50K100K150K200K250KSE +/- 192.59, N = 8SE +/- 323.17, N = 8SE +/- 242.07, N = 8SE +/- 390.91, N = 8SE +/- 1712.13, N = 15SE +/- 1894.96, N = 15SE +/- 1159.21, N = 15SE +/- 2579.89, N = 15SE +/- 2151.34, N = 15SE +/- 1538.33, N = 15SE +/- 4184.46, N = 15SE +/- 3403.37, N = 15SE +/- 667.76, N = 9SE +/- 1597.85, N = 15SE +/- 1067.93, N = 5SE +/- 1391.68, N = 15SE +/- 176.61, N = 8SE +/- 460.71, N = 9SE +/- 405.89, N = 9SE +/- 895.55, N = 9SE +/- 389.45, N = 9SE +/- 801.51, N = 863349.2964451.0263818.0963431.64125921.29225076.92118608.07193302.55195724.02137608.68219072.04235405.28127404.16246953.19249368.2567474.8656195.18113810.7288049.90162844.23107735.12217625.321. (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 4.1.5

LULESH

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

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P16K32K48K64K80KSE +/- 55.08, N = 3SE +/- 113.29, N = 3SE +/- 42.28, N = 3SE +/- 36.15, N = 3SE +/- 113.24, N = 3SE +/- 211.25, N = 3SE +/- 84.04, N = 3SE +/- 150.76, N = 3SE +/- 223.02, N = 3SE +/- 90.30, N = 3SE +/- 185.16, N = 3SE +/- 298.94, N = 3SE +/- 59.94, N = 3SE +/- 93.50, N = 3SE +/- 182.10, N = 3SE +/- 24.48, N = 5SE +/- 10.93, N = 5SE +/- 48.29, N = 4SE +/- 49.50, N = 5SE +/- 96.77, N = 4SE +/- 94.58, N = 4SE +/- 52.96, N = 416684.8016474.6416431.0616508.7724037.4446784.6823701.4445810.9146825.4224712.8847999.6849234.5922356.7543781.1944011.6518165.2418008.0133249.8623997.7245133.4539468.9172625.301. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

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 1.2EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1000M2000M3000M4000M5000MSE +/- 1287076.96, N = 3SE +/- 895904.07, N = 3SE +/- 1018274.19, N = 3SE +/- 346922.82, N = 3SE +/- 3180941.37, N = 3SE +/- 5545904.69, N = 3SE +/- 5502047.33, N = 3SE +/- 9413640.32, N = 3SE +/- 3198523.64, N = 3SE +/- 1953221.73, N = 3SE +/- 22527036.40, N = 3SE +/- 3930391.72, N = 3SE +/- 2811300.43, N = 3SE +/- 870457.03, N = 3SE +/- 2195794.19, N = 3SE +/- 2127099.93, N = 3SE +/- 321808.95, N = 3SE +/- 593246.81, N = 3SE +/- 269139.99, N = 3SE +/- 3825586.55, N = 3SE +/- 5689485.08, N = 3SE +/- 4788678.64, N = 311527833331154740667115066300011525826672321156000454522933322967306674514770333454912566723748533334659305667467521633322910496674531752333454989466714955353331079277000211994466716118260003174538667183991266737394853331. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsssp median_TEPS, More Is BetterGraph500 3.0Scale: 26EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P170M340M510M680M850M2679450002692330002494630002746570003680160004964920003911670005819650006560250003837540005897850006450760003770330007327990007779790001799770002214370002932990003331290003796550004191380004762740001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

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.CEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20K40K60K80K100KSE +/- 140.98, N = 6SE +/- 73.01, N = 6SE +/- 106.45, N = 6SE +/- 120.40, N = 6SE +/- 363.39, N = 8SE +/- 692.06, N = 8SE +/- 537.84, N = 15SE +/- 617.41, N = 11SE +/- 1124.02, N = 15SE +/- 469.70, N = 9SE +/- 704.17, N = 7SE +/- 760.49, N = 7SE +/- 391.92, N = 15SE +/- 452.76, N = 5SE +/- 451.86, N = 5SE +/- 51.94, N = 7SE +/- 38.35, N = 6SE +/- 34.92, N = 7SE +/- 268.67, N = 15SE +/- 478.84, N = 15SE +/- 536.89, N = 9SE +/- 755.83, N = 1022438.7222681.8222464.4322415.0445867.4485050.5854465.0485652.9585165.3759557.0881967.3987251.5644431.5866627.4767935.2632665.5722768.3639597.2135560.2561041.1566065.9696083.421. (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 4.1.5

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50K100K150K200K250KSE +/- 179.19, N = 6SE +/- 96.33, N = 6SE +/- 136.99, N = 6SE +/- 210.09, N = 6SE +/- 1162.69, N = 15SE +/- 1607.61, N = 15SE +/- 698.81, N = 8SE +/- 1530.10, N = 7SE +/- 1626.95, N = 15SE +/- 686.49, N = 8SE +/- 1670.64, N = 15SE +/- 1462.99, N = 7SE +/- 1190.42, N = 8SE +/- 1345.32, N = 5SE +/- 1737.88, N = 5SE +/- 1247.93, N = 15SE +/- 52.87, N = 6SE +/- 665.53, N = 7SE +/- 445.75, N = 6SE +/- 845.49, N = 15SE +/- 354.01, N = 7SE +/- 1707.04, N = 1568083.6268882.4767161.3968641.92124827.75243408.83125164.03231864.87232777.48124391.00229308.77242247.85141255.07243422.17249256.0658881.1459708.35100227.4870989.97118612.3398939.86181631.921. (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 4.1.5

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: SlowEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 4SE +/- 0.05, N = 4SE +/- 0.61, N = 5SE +/- 0.30, N = 5SE +/- 0.03, N = 4SE +/- 0.15, N = 5SE +/- 0.51, N = 5SE +/- 0.10, N = 4SE +/- 0.63, N = 15SE +/- 0.69, N = 15SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 424.1624.2322.8223.8329.2144.6239.4661.2463.1838.6361.9462.3938.1455.0355.9515.5715.2019.7520.2631.6424.3341.511. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbfs max_TEPS, More Is BetterGraph500 3.0Scale: 26EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400M800M1200M1600M2000M7565900007796100006939450007740870008651520001305120000839917000154399000017298000008821810001484490000159545000011845100002040450000202988000049103100080265900090414100011131200001094220000130420000012483300001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

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: SlowEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 4SE +/- 0.17, N = 4SE +/- 0.06, N = 3SE +/- 0.19, N = 4SE +/- 0.21, N = 4SE +/- 0.07, N = 3SE +/- 0.33, N = 15SE +/- 0.34, N = 7SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 317.6318.0516.3117.6821.7633.6128.5544.6944.6727.6041.9445.9329.2438.0338.1511.2511.0613.9515.8225.0615.9127.48

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 2023.2.devModel: Handwritten English Recognition FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.13, N = 3SE +/- 0.12, N = 3SE +/- 0.14, N = 3SE +/- 0.14, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.13, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.50, N = 12SE +/- 0.40, N = 3SE +/- 0.15, N = 3SE +/- 0.07, N = 3SE +/- 0.77, N = 3SE +/- 0.40, N = 3SE +/- 0.04, N = 3SE +/- 0.12, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 342.9638.2147.6943.9429.7432.3636.6839.1036.1537.4838.4433.7651.0245.3840.57101.0182.0180.2638.6827.2724.4124.861. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, 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 +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 310.799.6512.4812.187.444.369.695.274.949.095.474.9810.536.506.1611.4918.029.9916.8315.5911.9912.551. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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: MediumEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 4SE +/- 0.06, N = 4SE +/- 0.48, N = 5SE +/- 0.24, N = 5SE +/- 0.04, N = 4SE +/- 0.22, N = 5SE +/- 0.53, N = 5SE +/- 0.09, N = 4SE +/- 0.79, N = 15SE +/- 0.39, N = 12SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 424.7324.8823.4124.4229.9945.2940.0762.9663.9039.2361.5863.2238.8154.6554.9216.0315.5820.3821.0232.5024.7542.361. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 3.3Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.009380, N = 3SE +/- 0.009692, N = 3SE +/- 0.011111, N = 3SE +/- 0.018878, N = 3SE +/- 0.003964, N = 3SE +/- 0.011421, N = 3SE +/- 0.010405, N = 3SE +/- 0.014637, N = 3SE +/- 0.012212, N = 9SE +/- 0.002129, N = 3SE +/- 0.018763, N = 3SE +/- 0.016330, N = 3SE +/- 0.005750, N = 3SE +/- 0.022339, N = 3SE +/- 0.019644, N = 4SE +/- 0.006193, N = 3SE +/- 0.013679, N = 3SE +/- 0.007963, N = 3SE +/- 0.000209, N = 3SE +/- 0.002989, N = 15SE +/- 0.000663, N = 3SE +/- 0.003914, N = 31.5041901.3372001.7064101.5374301.1521101.2126201.2269801.4652101.4943801.2369701.5860001.5272301.2932801.8385001.8280505.6202306.6390103.9520300.2500140.4333220.2149600.3713531. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

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: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 15SE +/- 0.08, N = 15SE +/- 0.12, N = 3SE +/- 0.15, N = 15SE +/- 0.11, N = 6SE +/- 0.14, N = 15SE +/- 0.14, N = 15SE +/- 0.10, N = 15SE +/- 0.09, N = 7SE +/- 0.14, N = 15SE +/- 0.15, N = 15SE +/- 0.04, N = 3SE +/- 0.06, N = 15SE +/- 0.10, N = 3SE +/- 0.15, N = 12SE +/- 0.07, N = 3SE +/- 0.13, N = 157.848.607.018.2310.4810.4711.6010.4810.5811.3110.2710.6712.439.639.933.154.736.287.097.987.888.50

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbfs median_TEPS, More Is BetterGraph500 3.0Scale: 26EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400M800M1200M1600M2000M7233320007627280006566290007562890008361590001258070000816172000145658000016173400008540290001406670000148790000011470900001865120000183548000048058700078754500087443000010738600001051290000123867000011998300001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: SharpenEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30060090012001500SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 1.73, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 1.86, N = 3SE +/- 4.10, N = 3SE +/- 0.67, N = 3SE +/- 6.74, N = 3SE +/- 4.37, N = 3SE +/- 0.33, N = 3SE +/- 1.76, N = 3SE +/- 1.67, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 35305634845636691038809112213347751089131192412421328528351639691117374913021. (CC) gcc options: -fopenmp -O2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread

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: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.01, N = 3SE +/- 0.21, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.51, N = 15SE +/- 0.04, N = 3SE +/- 0.21, N = 3SE +/- 1.07, N = 15SE +/- 0.17, N = 15SE +/- 0.11, N = 3SE +/- 0.11, N = 3SE +/- 0.31, N = 15SE +/- 1.15, N = 15SE +/- 0.18, N = 3SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.17, N = 3SE +/- 0.10, N = 15SE +/- 0.19, N = 3SE +/- 0.07, N = 3SE +/- 0.31, N = 1514.5516.1713.2014.8018.5226.9021.1127.6433.5321.2326.3530.4224.6633.6032.1610.658.9814.2413.0917.1814.4719.54

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: MediumEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1122334455SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 4SE +/- 0.01, N = 3SE +/- 0.30, N = 4SE +/- 0.43, N = 4SE +/- 0.01, N = 3SE +/- 0.13, N = 4SE +/- 0.28, N = 4SE +/- 0.07, N = 3SE +/- 0.36, N = 4SE +/- 0.28, N = 15SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 319.9320.4818.3719.9624.6437.5032.0146.7945.2230.9443.2947.5332.6838.6138.9012.8912.8216.3017.9328.0116.7229.95

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: 60EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1632486480SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.18, N = 3SE +/- 0.35, N = 3SE +/- 0.05, N = 3SE +/- 0.54, N = 4SE +/- 0.28, N = 9SE +/- 0.15, N = 3SE +/- 0.52, N = 3SE +/- 0.27, N = 3SE +/- 0.09, N = 3SE +/- 0.69, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.11, N = 3SE +/- 0.56, N = 3SE +/- 0.08, N = 3SE +/- 0.33, N = 322.2422.3222.2522.3122.4549.0132.1343.9750.8623.7844.1173.2625.8947.0151.7528.9820.6540.3131.2460.4235.4270.951. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

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 2023.2.devModel: Age Gender Recognition Retail 0013 FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.2970.5940.8911.1881.485SE +/- 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 = 30.720.640.820.810.540.510.800.540.520.850.550.531.070.640.631.131.321.290.610.380.520.461. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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: Compression RatingEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P200K400K600K800K1000KSE +/- 329.36, N = 3SE +/- 120.37, N = 3SE +/- 88.26, N = 3SE +/- 110.87, N = 3SE +/- 224.17, N = 3SE +/- 2236.71, N = 3SE +/- 311.94, N = 3SE +/- 2398.87, N = 3SE +/- 3040.99, N = 3SE +/- 366.95, N = 3SE +/- 814.80, N = 3SE +/- 3010.91, N = 3SE +/- 439.85, N = 3SE +/- 1320.36, N = 3SE +/- 2373.22, N = 3SE +/- 2257.75, N = 3SE +/- 386.27, N = 3SE +/- 1114.34, N = 3SE +/- 629.56, N = 3SE +/- 3175.14, N = 3SE +/- 946.51, N = 3SE +/- 3381.54, N = 33543863647853328903570614897367340935649508098688610355544118020078534275865718271038509832919002482494119573896906041314487866705511. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

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.i3dEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9754EPYC 9754 2PXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120240360480600SE +/- 0.23, N = 3SE +/- 0.32, N = 3SE +/- 0.54, N = 3SE +/- 0.18, N = 3SE +/- 7.98, N = 9SE +/- 10.30, N = 11SE +/- 8.73, N = 9SE +/- 2.61, N = 9SE +/- 9.41, N = 9SE +/- 5.21, N = 9SE +/- 8.77, N = 12SE +/- 14.11, N = 3SE +/- 2.09, N = 8SE +/- 1.70, N = 3SE +/- 0.17, N = 3SE +/- 1.36, N = 3SE +/- 4.12, N = 4SE +/- 0.41, N = 3SE +/- 4.29, N = 3SE +/- 0.20, N = 3560.36555.30561.97556.88515.79235.75430.08247.18191.02440.55194.74493.50249.07444.93551.26289.90353.56193.57323.53167.091. (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

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: EnhancedEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400800120016002000SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.88, N = 3SE +/- 1.33, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 12.53, N = 3SE +/- 2.73, N = 3SE +/- 1.67, N = 3SE +/- 12.58, N = 3SE +/- 4.18, N = 3SE +/- 0.58, N = 3SE +/- 13.84, N = 3SE +/- 7.84, N = 3SE +/- 1.33, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 1.20, N = 3SE +/- 3.84, N = 390296883495111421600131916601934127716461931145118051879803612107711241763119219361. (CC) gcc options: -fopenmp -O2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lzstd -lm -lpthread

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 2023.2.devModel: Face Detection FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P140280420560700SE +/- 0.17, N = 3SE +/- 0.46, N = 3SE +/- 0.27, N = 3SE +/- 1.05, N = 3SE +/- 0.07, N = 3SE +/- 0.20, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.52, N = 3SE +/- 0.06, N = 3SE +/- 0.19, N = 3SE +/- 0.20, N = 3SE +/- 0.11, N = 3SE +/- 0.40, N = 3SE +/- 0.47, N = 3SE +/- 0.10, N = 3SE +/- 0.54, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.17, N = 3SE +/- 0.80, N = 3SE +/- 0.14, N = 3566.57475.33645.20597.72402.09206.80500.74250.24229.66529.30269.70232.52542.60271.81251.33204.74413.59206.28361.56288.82232.13237.361. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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: 512EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400M800M1200M1600M2000MSE +/- 174355.96, N = 3SE +/- 788106.03, N = 3SE +/- 249799.92, N = 3SE +/- 1894873.20, N = 3SE +/- 1233333.33, N = 3SE +/- 2367370.79, N = 3SE +/- 556776.44, N = 3SE +/- 1883554.56, N = 3SE +/- 3702401.62, N = 3SE +/- 2458319.21, N = 3SE +/- 1846016.73, N = 3SE +/- 3244140.02, N = 3SE +/- 1026861.45, N = 3SE +/- 2520802.34, N = 3SE +/- 1826046.12, N = 3SE +/- 734461.25, N = 3SE +/- 300240.64, N = 3SE +/- 722641.76, N = 3SE +/- 1999408.25, N = 3SE +/- 5342596.12, N = 3SE +/- 2861429.32, N = 3SE +/- 8712124.37, N = 38442100001004166667741760000923553333114523333317163666671172100000161916666716880666671142300000153573333316948666671384533333141833333314161333337104500005453333339692766679059800001209600000114693333314416333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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: NinjaEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300SE +/- 0.23, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.11, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 1.00, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.75, N = 3SE +/- 0.09, N = 3SE +/- 0.28, N = 3SE +/- 1.04, N = 5SE +/- 0.89, N = 3SE +/- 0.14, N = 3SE +/- 0.65, N = 3SE +/- 0.20, N = 3SE +/- 0.08, N = 3SE +/- 0.27, N = 3SE +/- 0.32, N = 3SE +/- 0.05, N = 3179.54168.03191.73173.86131.2090.33133.0791.0185.06131.9991.7585.45146.3998.9095.91218.51264.13159.18169.53110.86148.7496.47

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 2023.2.devModel: Handwritten English Recognition FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.19, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.79, N = 3SE +/- 8.87, N = 15SE +/- 0.64, N = 15SE +/- 0.11, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 351.5443.2658.7951.9335.5638.3144.5646.3141.7346.4148.0141.6648.6450.1746.5690.37110.0270.9048.7639.9938.2138.561. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P7K14K21K28K35KSE +/- 0.18, N = 3SE +/- 0.75, N = 3SE +/- 0.17, N = 3SE +/- 2.88, N = 3SE +/- 0.68, N = 3SE +/- 1.84, N = 3SE +/- 1.05, N = 3SE +/- 6.62, N = 3SE +/- 22.86, N = 3SE +/- 0.81, N = 3SE +/- 1.67, N = 3SE +/- 18.13, N = 3SE +/- 0.79, N = 3SE +/- 2.44, N = 3SE +/- 20.66, N = 3SE +/- 10.36, N = 3SE +/- 1.36, N = 3SE +/- 2.46, N = 3SE +/- 12.53, N = 3SE +/- 32.76, N = 3SE +/- 2.57, N = 3SE +/- 37.86, N = 32951.493455.342590.452723.424151.238037.054989.009948.8910749.674704.829459.5110682.546088.8512118.3912835.181839.121267.602425.379096.2627014.3116646.0132258.521. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

libxsmm

Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 256EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P14002800420056007000SE +/- 2.70, N = 3SE +/- 4.42, N = 3SE +/- 5.00, N = 3SE +/- 0.80, N = 3SE +/- 5.95, N = 3SE +/- 12.02, N = 3SE +/- 36.77, N = 3SE +/- 19.77, N = 3SE +/- 15.89, N = 3SE +/- 2.46, N = 3SE +/- 29.80, N = 3SE +/- 50.25, N = 3SE +/- 3.30, N = 3SE +/- 38.63, N = 3SE +/- 207.81, N = 6SE +/- 1.27, N = 3SE +/- 0.83, N = 3SE +/- 1.60, N = 3SE +/- 2.55, N = 3SE +/- 12.35, N = 3SE +/- 11.63, N = 3SE +/- 14.69, N = 31818.02028.41621.61653.82524.64643.42858.85647.36053.23147.85893.26303.03343.56236.95875.2450.5300.4600.21030.11575.91293.12274.51. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: allmodconfigEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P100200300400500SE +/- 0.61, N = 3SE +/- 0.67, N = 3SE +/- 0.89, N = 3SE +/- 0.51, N = 3SE +/- 0.61, N = 3SE +/- 1.76, N = 3SE +/- 0.41, N = 3SE +/- 0.19, N = 3SE +/- 1.03, N = 3SE +/- 0.50, N = 3SE +/- 0.26, N = 3SE +/- 0.80, N = 3SE +/- 2.48, N = 3SE +/- 1.07, N = 3SE +/- 0.62, N = 3SE +/- 0.54, N = 3SE +/- 1.19, N = 3SE +/- 0.53, N = 3SE +/- 0.60, N = 3SE +/- 0.93, N = 3SE +/- 0.45, N = 3SE +/- 0.79, N = 3330.02295.25348.48314.90238.29157.59284.25188.57174.24283.87186.85169.88319.43206.00212.98353.88451.85253.80289.97183.89245.08150.72

Intel Open Image Denoise

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.1Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-OnlyEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.10032.20063.30094.40125.5015SE +/- 0.00, N = 4SE +/- 0.00, N = 4SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 4SE +/- 0.01, N = 6SE +/- 0.00, N = 5SE +/- 0.01, N = 6SE +/- 0.02, N = 6SE +/- 0.00, N = 4SE +/- 0.01, N = 5SE +/- 0.02, N = 6SE +/- 0.00, N = 5SE +/- 0.02, N = 6SE +/- 0.02, N = 6SE +/- 0.00, N = 4SE +/- 0.00, N = 3SE +/- 0.00, N = 5SE +/- 0.00, N = 4SE +/- 0.01, N = 6SE +/- 0.00, N = 5SE +/- 0.01, N = 61.912.131.711.872.644.182.984.394.682.453.774.103.594.674.892.171.803.022.694.463.104.51

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 2.12Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.17, N = 15SE +/- 0.23, N = 3SE +/- 0.17, N = 518.9421.0016.9319.8524.7330.2027.0929.9532.0626.7529.0132.4028.6128.1328.2114.2511.8720.6420.0814.8822.6616.98

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 BuckyballEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9754EPYC 9754 2PXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P80016002400320040001955.41742.52189.11926.71452.11322.71568.41559.51440.11641.61614.61700.81721.42686.63505.42198.32056.61879.71850.81738.61. (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

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 171.05, N = 3SE +/- 169.08, N = 3SE +/- 97.88, N = 3SE +/- 104.77, N = 3SE +/- 104.66, N = 3SE +/- 2049.27, N = 3SE +/- 412.01, N = 3SE +/- 733.59, N = 3SE +/- 885.34, N = 3SE +/- 168.11, N = 3SE +/- 535.98, N = 3SE +/- 1726.60, N = 5SE +/- 250.75, N = 3SE +/- 1116.49, N = 3SE +/- 962.46, N = 3SE +/- 321.31, N = 3SE +/- 50.97, N = 3SE +/- 312.63, N = 3SE +/- 160.07, N = 3SE +/- 1854.77, N = 3SE +/- 214.75, N = 3SE +/- 535.52, N = 380817.2990317.6071656.6375320.93107563.74157704.81109112.78158898.75157993.59101207.73145836.62157084.79112191.74143642.66147460.16106600.2260464.31125095.63105652.29150083.4780814.50122410.811. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.1710.3420.5130.6840.855SE +/- 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 = 5SE +/- 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 = 30.530.480.610.590.400.340.600.340.340.650.360.340.760.430.430.460.550.440.480.290.440.391. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 19.8.1Time To CompileEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.13, N = 3SE +/- 0.32, N = 3SE +/- 0.32, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.56, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.47, N = 3SE +/- 0.17, N = 3SE +/- 0.08, N = 3SE +/- 0.26, N = 3SE +/- 0.29, N = 3SE +/- 0.53, N = 3SE +/- 1.00, N = 3SE +/- 0.23, N = 3SE +/- 0.27, N = 3SE +/- 0.02, N = 3SE +/- 0.13, N = 3SE +/- 0.12, N = 3165.10157.54174.55160.07121.9393.41120.1693.0589.41119.2092.3687.75130.46104.59103.88196.61229.91149.18151.57108.10134.0998.33

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 FastEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.04, N = 4SE +/- 0.04, N = 5SE +/- 0.01, N = 4SE +/- 0.03, N = 5SE +/- 0.06, N = 5SE +/- 0.42, N = 5SE +/- 0.13, N = 5SE +/- 0.13, N = 5SE +/- 0.35, N = 5SE +/- 0.09, N = 5SE +/- 0.39, N = 5SE +/- 0.34, N = 5SE +/- 0.34, N = 5SE +/- 0.58, N = 15SE +/- 0.45, N = 15SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.25, N = 4SE +/- 0.06, N = 4SE +/- 0.13, N = 5SE +/- 0.13, N = 3SE +/- 0.16, N = 552.3653.6648.2152.8863.5256.4063.4654.6953.0160.6255.2255.6755.1144.2746.1330.0930.5741.7849.7554.8535.8956.04

DuckDB

DuckDB is an in-progress SQL OLAP database management system optimized for analytics and features a vectorized and parallel engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDuckDB 0.9.1Benchmark: IMDBEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.14, N = 3SE +/- 0.16, N = 3SE +/- 1.14, N = 3SE +/- 0.18, N = 3SE +/- 1.32, N = 3SE +/- 0.84, N = 3SE +/- 0.26, N = 3SE +/- 0.68, N = 3SE +/- 1.46, N = 3SE +/- 0.08, N = 3SE +/- 0.51, N = 3SE +/- 1.54, N = 3SE +/- 0.10, N = 3SE +/- 0.02, N = 3SE +/- 0.12, N = 3SE +/- 0.24, N = 3SE +/- 0.08, N = 3SE +/- 0.50, N = 3SE +/- 0.39, N = 3123.61123.36126.96125.90103.72132.29115.81154.97152.23116.05154.19154.75147.60202.19202.07136.16110.73124.3199.25123.8396.87124.711. (CXX) g++ options: -O3 -rdynamic -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 6.1Build: defconfigEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1122334455SE +/- 0.54, N = 4SE +/- 0.41, N = 5SE +/- 0.49, N = 4SE +/- 0.50, N = 4SE +/- 0.39, N = 4SE +/- 0.21, N = 9SE +/- 0.46, N = 4SE +/- 0.18, N = 14SE +/- 0.22, N = 10SE +/- 0.45, N = 4SE +/- 0.22, N = 10SE +/- 0.21, N = 10SE +/- 0.44, N = 5SE +/- 0.28, N = 9SE +/- 0.26, N = 10SE +/- 0.52, N = 3SE +/- 0.42, N = 3SE +/- 0.38, N = 4SE +/- 0.49, N = 3SE +/- 0.27, N = 6SE +/- 0.37, N = 4SE +/- 0.22, N = 743.6641.3345.0342.3033.5226.8136.6028.8628.1236.5929.0827.8741.5533.5233.8541.5047.5633.6935.6027.6930.7723.78

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.36, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.26, N = 3SE +/- 0.03, N = 3SE +/- 0.30, N = 3SE +/- 0.23, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.18, N = 3SE +/- 0.20, N = 3SE +/- 0.09, N = 329.5034.2125.9827.8241.7982.6250.26102.59108.3548.0295.53108.5261.00121.93131.8418.6412.3624.6277.81217.11121.19237.831. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Tachyon

This is a test of the threaded Tachyon, a parallel ray-tracing system, measuring the time to ray-trace a sample scene. The sample scene used is the Teapot scene ray-traced to 8K x 8K with 32 samples. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.99.2Total TimeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.0186, N = 3SE +/- 0.0153, N = 3SE +/- 0.0309, N = 3SE +/- 0.0725, N = 3SE +/- 0.0261, N = 3SE +/- 0.0822, N = 4SE +/- 0.0078, N = 3SE +/- 0.0923, N = 5SE +/- 0.0556, N = 15SE +/- 0.0174, N = 3SE +/- 0.0475, N = 5SE +/- 0.0704, N = 15SE +/- 0.0160, N = 4SE +/- 0.0430, N = 6SE +/- 0.0644, N = 15SE +/- 0.7762, N = 3SE +/- 0.0600, N = 3SE +/- 0.0324, N = 3SE +/- 0.0435, N = 3SE +/- 0.0192, N = 4SE +/- 0.0503, N = 3SE +/- 0.0153, N = 427.658923.228332.089928.917020.323010.779317.16949.24118.022017.93369.44148.180313.26157.17336.9114128.855047.907024.715825.008612.976022.848211.95391. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P8001600240032004000SE +/- 2.61, N = 3SE +/- 0.75, N = 3SE +/- 3.62, N = 3SE +/- 1.15, N = 3SE +/- 13.47, N = 3SE +/- 1.36, N = 3SE +/- 34.13, N = 15SE +/- 0.25, N = 3SE +/- 4.43, N = 3SE +/- 2.17, N = 3SE +/- 0.51, N = 3SE +/- 1.11, N = 3SE +/- 19.21, N = 15SE +/- 31.11, N = 14SE +/- 15.99, N = 3SE +/- 2.73, N = 3SE +/- 0.25, N = 3SE +/- 0.34, N = 3SE +/- 0.15, N = 3SE +/- 1.25, N = 3SE +/- 1.11, N = 3SE +/- 4.07, N = 3583.02580.02570.01563.771317.722562.821262.823085.223435.761722.743097.023503.21668.593074.673769.74326.68225.49517.74746.671531.59986.421880.121. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2K4K6K8K10KSE +/- 0.18, N = 3SE +/- 0.90, N = 3SE +/- 0.56, N = 3SE +/- 1.04, N = 3SE +/- 31.36, N = 14SE +/- 1.72, N = 3SE +/- 35.78, N = 14SE +/- 3.26, N = 3SE +/- 10.10, N = 3SE +/- 1.34, N = 3SE +/- 1.87, N = 3SE +/- 10.51, N = 3SE +/- 25.68, N = 13SE +/- 79.68, N = 4SE +/- 61.28, N = 15SE +/- 2.67, N = 3SE +/- 0.73, N = 3SE +/- 1.30, N = 3SE +/- 0.73, N = 3SE +/- 1.87, N = 3SE +/- 3.42, N = 3SE +/- 3.34, N = 31427.131419.711426.191435.332458.346103.922645.577541.318081.043903.827470.148334.111502.826642.526561.39677.18545.331131.521874.943900.282440.054799.021. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 - GriddingEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 0.00, N = 2SE +/- 66.65, N = 2SE +/- 110.51, N = 3SE +/- 219.20, N = 3SE +/- 423.30, N = 3SE +/- 405.08, N = 3SE +/- 664.12, N = 3SE +/- 0.00, N = 3SE +/- 485.47, N = 3SE +/- 971.00, N = 3SE +/- 1162.33, N = 3SE +/- 655.18, N = 14SE +/- 839.32, N = 3SE +/- 336.69, N = 3SE +/- 25.00, N = 3SE +/- 66.98, N = 3SE +/- 86.95, N = 3SE +/- 146.74, N = 3SE +/- 163.80, N = 3SE +/- 227.17, N = 3SE +/- 423.30, N = 316660.0016726.7016190.4016928.8037265.9073231.6047003.2085111.70101573.0067965.50134960.00147615.0027269.2068286.4069975.4011749.209965.2819879.2022365.2043896.4037712.3072808.301. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 45.17, N = 3SE +/- 45.17, N = 3SE +/- 108.83, N = 3SE +/- 117.42, N = 3SE +/- 153.57, N = 3SE +/- 298.23, N = 3SE +/- 475.58, N = 3SE +/- 640.00, N = 3SE +/- 701.50, N = 3SE +/- 0.00, N = 3SE +/- 1159.96, N = 3SE +/- 856.67, N = 3SE +/- 574.27, N = 15SE +/- 305.39, N = 3SE +/- 183.50, N = 3SE +/- 18.19, N = 3SE +/- 53.97, N = 3SE +/- 39.13, N = 3SE +/- 131.23, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 494.81, N = 316838.5016838.5016066.5016750.3031177.2061441.9041081.6077438.7087476.5058310.10112479.00126807.0026593.4066643.0067898.5010018.818945.9617571.3019616.0038213.0030422.7059984.301. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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 3.3Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.9021.8042.7063.6084.51SE +/- 0.001247, N = 7SE +/- 0.001116, N = 7SE +/- 0.001918, N = 7SE +/- 0.001318, N = 7SE +/- 0.000411, N = 7SE +/- 0.002423, N = 7SE +/- 0.000826, N = 7SE +/- 0.002857, N = 15SE +/- 0.001307, N = 7SE +/- 0.002377, N = 7SE +/- 0.001854, N = 7SE +/- 0.003682, N = 15SE +/- 0.002327, N = 7SE +/- 0.003809, N = 15SE +/- 0.004528, N = 15SE +/- 0.001460, N = 7SE +/- 0.000729, N = 7SE +/- 0.000805, N = 7SE +/- 0.000549, N = 7SE +/- 0.002715, N = 7SE +/- 0.001490, N = 7SE +/- 0.002305, N = 70.5198050.4623580.6158840.5790510.3850510.3094310.3814130.3570810.3397650.3978690.3892510.3550310.3848440.4104650.4027103.1400404.0090902.0956800.3113360.3343530.2832500.2962341. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P12002400360048006000SE +/- 1.72, N = 3SE +/- 2.39, N = 3SE +/- 1.32, N = 3SE +/- 2.14, N = 3SE +/- 2.84, N = 3SE +/- 4.73, N = 3SE +/- 2.28, N = 3SE +/- 4.70, N = 3SE +/- 13.60, N = 3SE +/- 0.86, N = 3SE +/- 5.23, N = 3SE +/- 20.32, N = 3SE +/- 0.79, N = 3SE +/- 4.90, N = 3SE +/- 14.45, N = 3SE +/- 5.39, N = 3SE +/- 32.22, N = 15SE +/- 10.27, N = 15SE +/- 2.96, N = 3SE +/- 0.77, N = 3SE +/- 1.23, N = 3SE +/- 3.02, N = 31239.061475.851086.601229.831795.333333.342148.824138.944592.132064.473995.514605.882624.945097.395491.84619.26398.601128.841224.632998.941673.953317.931. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2004006008001000SE +/- 0.30, N = 3SE +/- 0.50, N = 3SE +/- 0.22, N = 3SE +/- 0.40, N = 3SE +/- 0.59, N = 3SE +/- 0.17, N = 3SE +/- 8.54, N = 15SE +/- 5.60, N = 15SE +/- 6.12, N = 14SE +/- 0.09, N = 3SE +/- 2.00, N = 3SE +/- 0.88, N = 3SE +/- 10.38, N = 15SE +/- 17.21, N = 15SE +/- 20.93, N = 15SE +/- 0.52, N = 3SE +/- 0.12, N = 3SE +/- 0.36, N = 3SE +/- 0.53, N = 3SE +/- 1.25, N = 3SE +/- 1.36, N = 3SE +/- 1.09, N = 3220.00216.77196.89215.71339.80667.62321.55794.35933.66476.94986.681105.95258.36762.57785.45105.0183.53175.66289.65638.73389.78751.311. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Detection FP32 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2004006008001000SE +/- 0.27, N = 3SE +/- 0.90, N = 3SE +/- 0.24, N = 3SE +/- 0.28, N = 3SE +/- 2.95, N = 15SE +/- 2.47, N = 3SE +/- 3.59, N = 14SE +/- 8.12, N = 3SE +/- 3.54, N = 3SE +/- 3.38, N = 11SE +/- 1.08, N = 3SE +/- 0.51, N = 3SE +/- 3.75, N = 13SE +/- 5.69, N = 15SE +/- 7.68, N = 14SE +/- 0.14, N = 3SE +/- 0.16, N = 3SE +/- 0.44, N = 3SE +/- 0.34, N = 3SE +/- 0.42, N = 3SE +/- 0.28, N = 3SE +/- 0.63, N = 3220.49218.19196.44215.69302.06667.11313.53796.46944.57448.72991.141105.04248.43750.98763.10104.4583.70174.35290.08639.86390.86751.241. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P14002800420056007000SE +/- 4.64, N = 3SE +/- 5.13, N = 3SE +/- 3.85, N = 3SE +/- 4.58, N = 3SE +/- 3.96, N = 3SE +/- 2.18, N = 3SE +/- 8.97, N = 3SE +/- 0.78, N = 3SE +/- 9.23, N = 3SE +/- 4.49, N = 3SE +/- 7.56, N = 3SE +/- 6.85, N = 3SE +/- 26.53, N = 12SE +/- 50.12, N = 3SE +/- 23.88, N = 3SE +/- 0.37, N = 3SE +/- 4.53, N = 3SE +/- 5.04, N = 3SE +/- 1.42, N = 3SE +/- 20.59, N = 3SE +/- 0.88, N = 3SE +/- 7.23, N = 31485.621668.821338.891452.182143.493944.922608.944899.575301.962553.244966.245660.382504.095633.746299.79553.91485.28996.051540.074392.732604.005110.001. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P6K12K18K24K30KSE +/- 1.57, N = 3SE +/- 6.03, N = 3SE +/- 10.97, N = 3SE +/- 17.68, N = 3SE +/- 10.36, N = 3SE +/- 13.82, N = 3SE +/- 15.74, N = 3SE +/- 7.50, N = 3SE +/- 11.09, N = 3SE +/- 9.91, N = 3SE +/- 6.07, N = 3SE +/- 15.44, N = 3SE +/- 14.67, N = 3SE +/- 7.25, N = 3SE +/- 42.44, N = 3SE +/- 0.26, N = 3SE +/- 3.13, N = 3SE +/- 5.36, N = 3SE +/- 0.55, N = 3SE +/- 11.79, N = 3SE +/- 2.90, N = 3SE +/- 9.59, N = 36884.017963.885697.825852.159442.9718148.5211192.6722196.2223914.3110693.4020657.2723435.3513857.8227396.2728813.743239.912357.534723.406108.8714919.939331.0618346.001. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection FP16 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30060090012001500SE +/- 0.33, N = 3SE +/- 1.19, N = 3SE +/- 0.09, N = 3SE +/- 5.83, N = 3SE +/- 0.35, N = 3SE +/- 0.18, N = 3SE +/- 0.38, N = 3SE +/- 0.09, N = 3SE +/- 1.40, N = 3SE +/- 0.11, N = 3SE +/- 0.09, N = 3SE +/- 0.89, N = 3SE +/- 0.11, N = 3SE +/- 1.76, N = 3SE +/- 0.95, N = 3SE +/- 3.26, N = 3SE +/- 8.82, N = 3SE +/- 1.06, N = 3SE +/- 0.34, N = 3SE +/- 0.12, N = 3SE +/- 0.22, N = 3SE +/- 0.04, N = 31080.15930.821226.111144.57763.58386.60951.06466.94441.63997.24501.23441.211043.50523.98483.45748.401604.96808.46384.69138.04131.89134.411. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 4.0Preset: ExhaustiveEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.0001, N = 3SE +/- 0.0018, N = 4SE +/- 0.0003, N = 3SE +/- 0.0078, N = 3SE +/- 0.0011, N = 4SE +/- 0.0038, N = 5SE +/- 0.0030, N = 4SE +/- 0.0173, N = 6SE +/- 0.0433, N = 6SE +/- 0.0025, N = 4SE +/- 0.0099, N = 6SE +/- 0.0433, N = 6SE +/- 0.0013, N = 5SE +/- 0.0473, N = 6SE +/- 0.0196, N = 6SE +/- 0.0026, N = 3SE +/- 0.0033, N = 3SE +/- 0.0038, N = 3SE +/- 0.0013, N = 4SE +/- 0.0556, N = 5SE +/- 0.0019, N = 4SE +/- 0.0084, N = 53.88524.45183.41563.76135.02529.55755.960511.703613.69026.051411.908713.48658.040215.816116.76062.11131.48832.97304.10948.06074.56378.85281. (CXX) g++ options: -O3 -flto -pthread

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120240360480600SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.14, N = 3SE +/- 0.02, N = 3SE +/- 0.15, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 3SE +/- 0.54, N = 3SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.02, N = 3SE +/- 0.40, N = 3SE +/- 0.52, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.25, N = 3SE +/- 0.74, N = 3SE +/- 0.45, N = 356.2867.0849.4253.3379.47154.4895.65191.33208.8290.46177.55206.06117.46235.00254.0568.3148.2496.82165.49414.77275.46538.201. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: Triple SHA-256, OnecoinEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P200K400K600K800K1000KSE +/- 386.45, N = 3SE +/- 57.83, N = 3SE +/- 136.14, N = 3SE +/- 2092.23, N = 13SE +/- 368.89, N = 3SE +/- 6450.09, N = 5SE +/- 2695.65, N = 3SE +/- 2730.73, N = 3SE +/- 1452.27, N = 3SE +/- 301.68, N = 3SE +/- 1477.55, N = 3SE +/- 703.16, N = 3SE +/- 260.26, N = 3SE +/- 3827.83, N = 3SE +/- 6976.69, N = 3SE +/- 712.00, N = 3SE +/- 69.84, N = 3SE +/- 65.57, N = 3SE +/- 8.82, N = 3SE +/- 833.51, N = 3SE +/- 586.01, N = 3SE +/- 14.53, N = 32350472689432075302444853137636211523809277498138698133729937413578523804951509945801047963142203941531843001945833851832088334139371. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P10K20K30K40K50KSE +/- 0.38, N = 3SE +/- 2.42, N = 3SE +/- 1.87, N = 3SE +/- 6.97, N = 3SE +/- 2.09, N = 3SE +/- 17.63, N = 3SE +/- 3.66, N = 3SE +/- 11.34, N = 3SE +/- 21.11, N = 3SE +/- 0.61, N = 3SE +/- 11.75, N = 3SE +/- 21.91, N = 3SE +/- 1.99, N = 3SE +/- 18.50, N = 3SE +/- 17.57, N = 3SE +/- 2.61, N = 3SE +/- 2.62, N = 3SE +/- 4.84, N = 3SE +/- 8.21, N = 3SE +/- 41.81, N = 3SE +/- 29.60, N = 3SE +/- 394.72, N = 35623.016751.674878.405291.707920.4515810.509559.3719365.6721326.438926.3318004.3620993.0911729.4523307.1525362.346817.314520.019727.1516507.1347285.1926776.4549016.721. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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.CEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P80K160K240K320K400KSE +/- 37.73, N = 3SE +/- 39.19, N = 3SE +/- 29.05, N = 3SE +/- 63.30, N = 3SE +/- 359.87, N = 4SE +/- 1114.28, N = 6SE +/- 250.78, N = 4SE +/- 1627.28, N = 5SE +/- 1006.53, N = 5SE +/- 1294.93, N = 6SE +/- 3065.86, N = 5SE +/- 3417.77, N = 7SE +/- 303.64, N = 4SE +/- 1784.74, N = 9SE +/- 931.89, N = 4SE +/- 169.72, N = 3SE +/- 22.16, N = 3SE +/- 135.49, N = 3SE +/- 69.75, N = 3SE +/- 171.62, N = 4SE +/- 105.41, N = 4SE +/- 1670.75, N = 552624.7153199.7253144.2153304.52119472.30242455.07125373.87263543.40272752.37208538.32353106.44390558.06131257.15218198.49224196.7649939.5637550.9690836.7872703.35152640.65111687.48242574.381. (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 4.1.5

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 4.0Preset: ThoroughEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.02, N = 5SE +/- 0.03, N = 5SE +/- 0.01, N = 4SE +/- 0.04, N = 4SE +/- 0.01, N = 5SE +/- 0.04, N = 6SE +/- 0.01, N = 6SE +/- 0.13, N = 6SE +/- 0.07, N = 7SE +/- 0.01, N = 6SE +/- 0.06, N = 6SE +/- 0.12, N = 7SE +/- 0.02, N = 6SE +/- 0.13, N = 6SE +/- 0.16, N = 6SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 4SE +/- 0.01, N = 5SE +/- 0.03, N = 6SE +/- 0.02, N = 5SE +/- 0.07, N = 637.9142.1132.4135.0947.6090.0856.41105.48120.0356.47104.88116.9974.35130.64138.3818.9113.5926.8642.3379.1546.4687.951. (CXX) g++ options: -O3 -flto -pthread

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 CompileEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.17, N = 3SE +/- 0.18, N = 3SE +/- 0.40, N = 3SE +/- 0.22, N = 3SE +/- 0.03, N = 3SE +/- 0.23, N = 3SE +/- 0.16, N = 3SE +/- 0.07, N = 3SE +/- 0.42, N = 3SE +/- 0.06, N = 3SE +/- 0.23, N = 3SE +/- 0.26, N = 3SE +/- 0.26, N = 3SE +/- 1.01, N = 3SE +/- 0.69, N = 15SE +/- 0.65, N = 3SE +/- 0.64, N = 3SE +/- 1.05, N = 3SE +/- 0.43, N = 3SE +/- 0.47, N = 3SE +/- 0.68, N = 3SE +/- 0.18, N = 3137.47131.92142.99133.95103.5092.98101.5692.4390.8898.5689.6287.17118.25110.07110.10160.90169.30131.14128.47109.56111.9693.30

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.7Encoder Mode: Preset 8 - Input: Bosphorus 4KEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.62, N = 15SE +/- 0.93, N = 5SE +/- 0.43, N = 5SE +/- 0.88, N = 6SE +/- 0.27, N = 6SE +/- 0.99, N = 15SE +/- 1.18, N = 15SE +/- 0.63, N = 5SE +/- 1.14, N = 15SE +/- 1.13, N = 15SE +/- 0.98, N = 5SE +/- 1.26, N = 15SE +/- 0.68, N = 15SE +/- 0.96, N = 15SE +/- 0.92, N = 15SE +/- 0.37, N = 4SE +/- 0.17, N = 5SE +/- 0.57, N = 5SE +/- 0.28, N = 5SE +/- 0.64, N = 15SE +/- 0.47, N = 5SE +/- 0.32, N = 590.5291.6784.6189.28107.90101.30110.29109.70109.36107.90106.92109.8397.9898.9699.4859.9974.6979.1890.0780.6296.9984.351. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

RawTherapee

RawTherapee is a cross-platform, open-source multi-threaded RAW image processing program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark TimeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.21, N = 3SE +/- 0.19, N = 3SE +/- 0.03, N = 3SE +/- 0.14, N = 3SE +/- 0.22, N = 3SE +/- 0.22, N = 3SE +/- 0.01, N = 3SE +/- 0.18, N = 3SE +/- 0.14, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.13, N = 359.4659.4860.3259.6249.9458.3352.0564.6864.9352.9965.0064.3666.1381.0281.3359.5953.3855.5346.5853.6745.5357.791. RawTherapee, version 5.9, command line.

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 CompileEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 2.95, N = 3SE +/- 1.69, N = 12SE +/- 2.79, N = 3SE +/- 2.31, N = 3SE +/- 1.67, N = 3SE +/- 1.28, N = 3SE +/- 1.09, N = 3SE +/- 1.37, N = 3SE +/- 0.56, N = 3SE +/- 1.97, N = 3SE +/- 1.57, N = 3SE +/- 0.78, N = 3SE +/- 2.85, N = 3SE +/- 0.96, N = 3SE +/- 1.84, N = 3SE +/- 2.67, N = 4SE +/- 2.78, N = 5SE +/- 2.01, N = 3SE +/- 1.53, N = 3SE +/- 0.26, N = 3SE +/- 1.77, N = 6SE +/- 1.46, N = 3222.76217.52231.49218.06179.74156.21180.86162.87153.27180.06159.58157.52208.58185.75182.75235.87251.63193.02197.76161.44174.18146.62

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.7Encoder Mode: Preset 4 - Input: Bosphorus 4KEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.32122.64243.96365.28486.606SE +/- 0.023, N = 3SE +/- 0.019, N = 3SE +/- 0.008, N = 3SE +/- 0.026, N = 3SE +/- 0.006, N = 3SE +/- 0.017, N = 3SE +/- 0.025, N = 3SE +/- 0.006, N = 3SE +/- 0.013, N = 3SE +/- 0.030, N = 3SE +/- 0.011, N = 3SE +/- 0.018, N = 3SE +/- 0.006, N = 3SE +/- 0.008, N = 3SE +/- 0.008, N = 3SE +/- 0.007, N = 3SE +/- 0.019, N = 3SE +/- 0.018, N = 3SE +/- 0.044, N = 3SE +/- 0.041, N = 3SE +/- 0.038, N = 3SE +/- 0.061, N = 34.8254.8104.7854.8045.8145.7485.8725.8075.8065.8245.7705.8294.9434.8994.9213.4874.0634.0894.9304.7705.6355.3511. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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 2023.2.devModel: Face Detection Retail FP16-INT8 - Device: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, 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 +/- 0.00, 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 = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 36.415.367.527.094.564.755.775.865.286.176.105.296.206.425.996.736.485.906.285.685.175.151. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv

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 21.0Time To CompileEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P714212835SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.13, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.22, N = 3SE +/- 0.22, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 324.3323.5525.0424.0219.4018.4619.9218.9618.8220.0819.1718.7823.5922.3422.4728.3428.4524.9622.6720.6720.0118.23

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.7Encoder Mode: Preset 13 - Input: Bosphorus 4KEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.78, N = 6SE +/- 0.97, N = 6SE +/- 1.56, N = 6SE +/- 0.77, N = 6SE +/- 1.69, N = 7SE +/- 3.33, N = 15SE +/- 0.55, N = 7SE +/- 1.37, N = 11SE +/- 0.70, N = 7SE +/- 0.21, N = 7SE +/- 1.49, N = 8SE +/- 0.95, N = 6SE +/- 0.73, N = 6SE +/- 1.57, N = 15SE +/- 1.57, N = 15SE +/- 0.60, N = 6SE +/- 0.29, N = 7SE +/- 1.46, N = 6SE +/- 0.58, N = 7SE +/- 1.38, N = 7SE +/- 0.32, N = 8SE +/- 1.07, N = 7186.04187.76178.67186.73201.39181.70190.13187.85190.76180.68172.71178.47177.55179.23177.46135.32165.20172.12194.18187.32210.90197.091. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.7Encoder Mode: Preset 12 - Input: Bosphorus 4KEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.94, N = 6SE +/- 0.35, N = 6SE +/- 0.37, N = 6SE +/- 1.86, N = 6SE +/- 1.64, N = 7SE +/- 3.24, N = 15SE +/- 0.64, N = 7SE +/- 1.51, N = 7SE +/- 1.58, N = 7SE +/- 0.23, N = 7SE +/- 0.94, N = 6SE +/- 1.15, N = 15SE +/- 1.54, N = 6SE +/- 1.96, N = 15SE +/- 1.96, N = 12SE +/- 0.50, N = 6SE +/- 0.26, N = 7SE +/- 1.23, N = 7SE +/- 0.71, N = 7SE +/- 0.87, N = 7SE +/- 0.48, N = 8SE +/- 1.29, N = 7186.12187.51182.15184.35200.85176.69193.26190.35189.57181.37175.72178.49178.60178.52175.56135.83165.06172.88194.62185.99209.10193.111. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read WriteEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20K40K60K80K100KSE +/- 457.09, N = 3SE +/- 501.89, N = 3SE +/- 467.18, N = 3SE +/- 129.67, N = 3SE +/- 447.56, N = 3SE +/- 264.40, N = 3SE +/- 206.72, N = 3SE +/- 298.54, N = 3SE +/- 575.17, N = 3SE +/- 72.21, N = 3SE +/- 232.38, N = 3SE +/- 673.42, N = 4SE +/- 149.60, N = 3SE +/- 555.15, N = 12SE +/- 364.36, N = 3SE +/- 177.69, N = 3SE +/- 433.25, N = 12SE +/- 167.75, N = 3SE +/- 79.20, N = 3SE +/- 204.55, N = 3SE +/- 222.67, N = 3SE +/- 446.45, N = 3619166224457178618687286060352782125717457589676065465755986714835755756626567156476971624706046756575232638521. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average LatencyEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.14, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.17, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.21, N = 4SE +/- 0.03, N = 3SE +/- 0.16, N = 12SE +/- 0.11, N = 3SE +/- 0.06, N = 3SE +/- 0.11, N = 12SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 316.1516.0717.4916.1613.7316.5712.7917.4917.3714.7918.3017.8713.9917.3917.6617.6315.4513.9614.1614.8013.2915.661. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

DuckDB

DuckDB is an in-progress SQL OLAP database management system optimized for analytics and features a vectorized and parallel engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDuckDB 0.9.1Benchmark: TPC-H ParquetEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.16, N = 3SE +/- 0.13, N = 3SE +/- 0.39, N = 3SE +/- 0.12, N = 3SE +/- 0.53, N = 3SE +/- 0.19, N = 3SE +/- 0.47, N = 3SE +/- 0.23, N = 3SE +/- 0.86, N = 3SE +/- 0.26, N = 3SE +/- 0.53, N = 3SE +/- 0.58, N = 3SE +/- 0.64, N = 3SE +/- 0.76, N = 3SE +/- 0.18, N = 3SE +/- 0.28, N = 3SE +/- 0.28, N = 3SE +/- 0.24, N = 3SE +/- 0.30, N = 3SE +/- 0.64, N = 3SE +/- 0.24, N = 3SE +/- 0.42, N = 3171.66171.73172.20171.49143.31148.52146.28156.70157.77147.49158.48156.16177.13191.65190.64177.76167.22169.57148.10156.79134.73149.981. (CXX) g++ options: -O3 -rdynamic -lssl -lcrypto -ldl

CPU Power Consumption Monitor

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30060090012001500Min: 6.69 / Avg: 129.67 / Max: 202.49Min: 7.41 / Avg: 152.38 / Max: 231.45Min: 7.38 / Avg: 109.07 / Max: 151.68Min: 7.36 / Avg: 129.87 / Max: 180.32Min: 11.14 / Avg: 227.12 / Max: 369.31Min: 25.64 / Avg: 377.42 / Max: 1445.04Min: 13.35 / Avg: 244.91 / Max: 809.65Min: 27.7 / Avg: 450.97 / Max: 795.91Min: 41.73 / Avg: 535.39 / Max: 826.13Min: 24.82 / Avg: 251.58 / Max: 451.13Min: 52.3 / Avg: 531.9 / Max: 682.51Min: 31.29 / Avg: 553.75 / Max: 832.25Min: 18.66 / Avg: 231.83 / Max: 866.74Min: 20.79 / Avg: 405.86 / Max: 1684.44Min: 20.73 / Avg: 473.08 / Max: 824.52Min: 17.45 / Avg: 306.45 / Max: 469.62Min: 28.7 / Avg: 236.79 / Max: 291.83Min: 58.7 / Avg: 448.7 / Max: 584.15Min: 60.21 / Avg: 309.16 / Max: 380.81Min: 144.84 / Avg: 559.17 / Max: 969.19Min: 37.56 / Avg: 289.52 / Max: 434.36Min: 100.34 / Avg: 556.83 / Max: 836.11

Stress-NG

OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption MonitorEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P130260390520650Min: 13.44 / Avg: 111.53 / Max: 125.64Min: 14.87 / Avg: 112.16 / Max: 133.12Min: 13.52 / Avg: 106.25 / Max: 121.24Min: 13.87 / Avg: 109.2 / Max: 124.09Min: 22.37 / Avg: 207.66 / Max: 237.33Min: 44.05 / Avg: 414.31 / Max: 533.24Min: 26.86 / Avg: 248.16 / Max: 285.38Min: 55.62 / Avg: 484.45 / Max: 608.08Min: 51.27 / Avg: 510.6 / Max: 675.99Min: 31.12 / Avg: 261.55 / Max: 298.34Min: 63.39 / Avg: 505.73 / Max: 619.28Min: 55.09 / Avg: 594.76 / Max: 754.93Min: 78.35 / Avg: 189.96 / Max: 248.19Min: 37.63 / Avg: 418.98 / Max: 577.71Min: 34.98 / Avg: 425.07 / Max: 594.67Min: 34.93 / Avg: 297.88 / Max: 337.93Min: 55.58 / Avg: 244.01 / Max: 273.95Min: 111.16 / Avg: 488.56 / Max: 562.14Min: 74.01 / Avg: 317.75 / Max: 377.23Min: 163.25 / Avg: 632 / Max: 756.86Min: 75.49 / Avg: 316.85 / Max: 377.02Min: 198.83 / Avg: 633.34 / Max: 755.49

OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: Matrix 3D MathEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P153045607553.8453.5056.3454.9065.1146.9254.1938.0538.9063.6044.4142.0342.1642.7344.3423.1726.3826.1931.1428.9743.7340.24

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Matrix 3D MathEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P5K10K15K20K25KSE +/- 0.87, N = 3SE +/- 3.44, N = 3SE +/- 2.25, N = 3SE +/- 2.70, N = 3SE +/- 1.74, N = 3SE +/- 806.67, N = 15SE +/- 8.72, N = 3SE +/- 851.10, N = 15SE +/- 951.59, N = 12SE +/- 224.21, N = 3SE +/- 1421.25, N = 12SE +/- 1251.58, N = 12SE +/- 776.60, N = 12SE +/- 770.85, N = 15SE +/- 924.24, N = 12SE +/- 4.31, N = 3SE +/- 2.22, N = 3SE +/- 19.78, N = 3SE +/- 40.38, N = 3SE +/- 47.17, N = 3SE +/- 12.58, N = 3SE +/- 55.08, N = 36004.426000.685986.415994.3613521.3419439.8913448.2018432.2919863.3016634.4422460.1924999.428009.2117903.5818848.546902.716436.6112793.849894.2018308.8413854.3825487.421. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Stockfish

OpenBenchmarking.orgWatts, Fewer Is BetterStockfish 15CPU Power Consumption MonitorEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P150300450600750Min: 14.3 / Avg: 164.37 / Max: 187.81Min: 14.29 / Avg: 212.74 / Max: 231.22Min: 13.67 / Avg: 129.62 / Max: 140.6Min: 14.02 / Avg: 158.87 / Max: 177.24Min: 22.74 / Avg: 318.59 / Max: 349.22Min: 47.65 / Avg: 612.65 / Max: 692.16Min: 26.6 / Avg: 320.13 / Max: 357.31Min: 57.84 / Avg: 643.53 / Max: 721.6Min: 57.78 / Avg: 732.85 / Max: 827.8Min: 31.01 / Avg: 312.39 / Max: 349.64Min: 62.72 / Avg: 628.97 / Max: 723.04Min: 63.28 / Avg: 746.49 / Max: 825.96Min: 80.5 / Avg: 322.28 / Max: 354.03Min: 39.73 / Avg: 622.05 / Max: 711.92Min: 39.44 / Avg: 698.81 / Max: 823.31Min: 34.83 / Avg: 371.53 / Max: 425.05Min: 55.09 / Avg: 260.12 / Max: 290.81Min: 113.33 / Avg: 518.05 / Max: 582.97Min: 57.15 / Avg: 334.51 / Max: 377.08Min: 162.35 / Avg: 669.46 / Max: 753Min: 75.78 / Avg: 333.36 / Max: 378.11Min: 198.72 / Avg: 662.32 / Max: 755.46

OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterStockfish 15Total TimeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P300K600K900K1200K1500K1076619.52980835.801184949.841083501.35758877.46673446.74902959.90794623.80732061.96902665.89762300.73720001.071127366.71934263.68891203.53334535.61347783.02333002.09458790.97411332.41502740.81449234.95

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 15Total TimeEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P130M260M390M520M650MSE +/- 1621191.42, N = 3SE +/- 2264497.13, N = 3SE +/- 2449641.07, N = 15SE +/- 3195340.10, N = 15SE +/- 3637655.15, N = 15SE +/- 841626.11, N = 3SE +/- 2088420.80, N = 3SE +/- 7826238.60, N = 15SE +/- 5692802.94, N = 3SE +/- 5369723.06, N = 15SE +/- 5080316.55, N = 12SE +/- 6642485.08, N = 15SE +/- 5395369.85, N = 15SE +/- 6629046.33, N = 15SE +/- 7837663.41, N = 15SE +/- 1984054.26, N = 15SE +/- 1454764.24, N = 15SE +/- 1857640.39, N = 15SE +/- 3106997.57, N = 15SE +/- 5058162.99, N = 15SE +/- 3048564.26, N = 15SE +/- 3459585.76, N = 4176962395208664442153597574172132834241774328412586511289066909511362452536489026281982711479462631537473172363329642581155326622781701124291134904652431725109981534692652753716011675942902975350721. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto=jobserver

LuxCoreRender

OpenBenchmarking.orgWatts, Fewer Is BetterLuxCoreRender 2.6CPU Power Consumption MonitorEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P150300450600750Min: 14.89 / Avg: 159 / Max: 176Min: 14.13 / Avg: 209.75 / Max: 230.65Min: 13.92 / Avg: 122.04 / Max: 141.21Min: 14.07 / Avg: 158.41 / Max: 174.88Min: 21.43 / Avg: 291.97 / Max: 316.58Min: 41.49 / Avg: 528.9 / Max: 633.51Min: 27.82 / Avg: 297.6 / Max: 323.2Min: 55.26 / Avg: 577.93 / Max: 658.14Min: 58.42 / Avg: 684.69 / Max: 822.99Min: 31.01 / Avg: 300 / Max: 325.32Min: 63.46 / Avg: 584.69 / Max: 655.72Min: 62.9 / Avg: 725.58 / Max: 832.25Min: 79.05 / Avg: 306.58 / Max: 334.14Min: 40.62 / Avg: 541.55 / Max: 666.9Min: 36.82 / Avg: 535.11 / Max: 803.08Min: 34.38 / Avg: 349.74 / Max: 386.57Min: 56.47 / Avg: 254.66 / Max: 289.67Min: 114.68 / Avg: 512.53 / Max: 580.05Min: 64.91 / Avg: 330.67 / Max: 375.27Min: 157.68 / Avg: 663.91 / Max: 752.74Min: 75.13 / Avg: 329.97 / Max: 376.94Min: 194.91 / Avg: 662.64 / Max: 752.3

OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.0160.0320.0480.0640.080.0610.0540.0710.0640.0450.0320.0490.0330.0320.0470.0390.0300.0550.0390.0380.0190.0220.0180.0260.0160.0280.019

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: CPUEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.01, N = 3SE +/- 0.09, N = 9SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.19, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.59, N = 15SE +/- 0.86, N = 15SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.82, N = 15SE +/- 0.19, N = 3SE +/- 0.95, N = 15SE +/- 0.54, N = 15SE +/- 0.06, N = 15SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.20, N = 15SE +/- 0.13, N = 3SE +/- 0.22, N = 159.6711.338.6410.1013.0416.7914.4819.0922.0514.0622.7321.4916.7621.2420.436.545.609.058.5210.889.4012.41

John The Ripper

OpenBenchmarking.orgWatts, Fewer Is BetterJohn The Ripper 2023.03.14CPU Power Consumption MonitorEPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P140280420560700Min: 13.75 / Avg: 159.45 / Max: 180.83Min: 7.44 / Avg: 199.79 / Max: 230.36Min: 13.66 / Avg: 129.04 / Max: 142.74Min: 13.77 / Avg: 144.34 / Max: 167.08Min: 22.12 / Avg: 308.92 / Max: 356.54Min: 43.78 / Avg: 542 / Max: 711.54Min: 27.11 / Avg: 308.53 / Max: 358.3Min: 56.62 / Avg: 565.76 / Max: 696.69Min: 57.91 / Avg: 634.46 / Max: 804.19Min: 29.14 / Avg: 297.38 / Max: 338.56Min: 62.1 / Avg: 541.22 / Max: 670.81Min: 63.28 / Avg: 651.59 / Max: 810.78Min: 78.5 / Avg: 274.72 / Max: 358.34Min: 38.92 / Avg: 501.54 / Max: 696.47Min: 40.16 / Avg: 568.7 / Max: 806.9Min: 34.3 / Avg: 381.35 / Max: 456.2Min: 54.98 / Avg: 255.8 / Max: 290.43Min: 107.78 / Avg: 508.54 / Max: 581.33Min: 67.85 / Avg: 330.72 / Max: 377.04Min: 154.39 / Avg: 654.68 / Max: 751.44Min: 75.06 / Avg: 328.59 / Max: 376.32Min: 195.58 / Avg: 627.72 / Max: 753.9

OpenBenchmarking.orgReal C/S Per Watt, More Is BetterJohn The Ripper 2023.03.14Test: MD5EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P14K28K42K56K70K59675.6554483.2465140.3860520.2542662.2240795.8551091.3947994.0445251.0049767.4445104.7544144.8052624.0462474.3059363.0719858.7721405.9219093.5528959.2826745.5230299.5830352.81

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: MD5EPYC 8534PEPYC 8534P - PowerEPYC 8534PNEPYC 8534PN - PowerEPYC 9554EPYC 9554 2PEPYC 9654EPYC 9654 2PEPYC 9654 2P - PowerEPYC 9684XEPYC 9684X 2PEPYC 9684X 2P - PowerEPYC 9754EPYC 9754 2PEPYC 9754 2P - PowerXeon Platinum 8280 2PXeon Platinum 8380Xeon Platinum 8380 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P7M14M21M28M35MSE +/- 4509.25, N = 3SE +/- 8987.65, N = 3SE +/- 3480.10, N = 3SE +/- 35713.37, N = 3SE +/- 3179.80, N = 3SE +/- 748908.30, N = 15SE +/- 3214.55, N = 3SE +/- 276676.91, N = 3SE +/- 126918.61, N = 3SE +/- 55480.73, N = 3SE +/- 527651.10, N = 15SE +/- 80354.08, N = 3SE +/- 835422.69, N = 15SE +/- 408087.55, N = 3SE +/- 740463.67, N = 15SE +/- 159107.00, N = 15SE +/- 7172.48, N = 3SE +/- 272625.73, N = 12SE +/- 10493.38, N = 3SE +/- 184527.95, N = 4SE +/- 214481.47, N = 15SE +/- 142430.10, N = 39515000108853338405667873566713179333221114001576300027153333287100001479966724411733287643331445666731333333337599337573133547566797097509577333175097509956267190530001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

158 Results Shown

OpenVINO
Stress-NG
Cpuminer-Opt:
  Deepcoin
  Myriad-Groestl
Stress-NG:
  AVX-512 VNNI
  Vector Floating Point
Xmrig
OpenVINO
7-Zip Compression
OpenVINO:
  Machine Translation EN To DE FP16 - CPU
  Vehicle Detection FP16 - CPU
Primesieve
Cpuminer-Opt
Blender
OpenVINO
Stress-NG
John The Ripper
Stress-NG
John The Ripper
Blender
miniBUDE:
  OpenMP - BM2:
    GFInst/s
    Billion Interactions/s
Blender
Coremark
OpenRadioss
OSPRay Studio:
  1 - 4K - 16 - Path Tracer - CPU
  2 - 4K - 16 - Path Tracer - CPU
OpenVINO:
  Face Detection Retail FP16-INT8 - CPU
  Road Segmentation ADAS FP16-INT8 - CPU
OSPRay Studio
Stress-NG:
  CPU Stress
  Floating Point
Blender
OSPRay Studio:
  1 - 4K - 32 - Path Tracer - CPU
  2 - 4K - 32 - Path Tracer - CPU
  3 - 4K - 32 - Path Tracer - CPU
OpenVINO
ACES DGEMM
OSPRay Studio
libxsmm
Stress-NG
OSPRay Studio:
  3 - 4K - 1 - Path Tracer - CPU
  2 - 4K - 1 - Path Tracer - CPU
OpenVINO
NAS Parallel Benchmarks
Blender
OpenVINO
NAS Parallel Benchmarks
Xmrig
Helsing
NAS Parallel Benchmarks
Stress-NG
Xcompact3d Incompact3d
John The Ripper
rays1bench
OpenVINO
OSPRay:
  gravity_spheres_volume/dim_512/ao/real_time
  gravity_spheres_volume/dim_512/scivis/real_time
OpenVINO
OSPRay
OpenVINO
OSPRay
OpenVINO:
  Face Detection Retail FP16 - CPU
  Weld Porosity Detection FP16-INT8 - CPU
NAMD
Embree
Rodinia
ASTC Encoder
Liquid-DSP
OpenVKL
Graph500
Embree
asmFish
OpenVINO
GROMACS
NAS Parallel Benchmarks
LULESH
Algebraic Multi-Grid Benchmark
Graph500
NAS Parallel Benchmarks:
  CG.C
  FT.C
Kvazaar
Graph500
uvg266
OpenVINO:
  Handwritten English Recognition FP16 - CPU
  Person Vehicle Bike Detection FP16 - CPU
Kvazaar
oneDNN
LuxCoreRender
Graph500
GraphicsMagick
LuxCoreRender
uvg266
High Performance Conjugate Gradient
OpenVINO
7-Zip Compression
Xcompact3d Incompact3d
GraphicsMagick
OpenVINO
Liquid-DSP
Timed LLVM Compilation
OpenVINO:
  Handwritten English Recognition FP16-INT8 - CPU
  Weld Porosity Detection FP16 - CPU
libxsmm
Timed Linux Kernel Compilation
Intel Open Image Denoise
OSPRay
NWChem
OpenVINO:
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU:
    FPS
    ms
Timed Node.js Compilation
uvg266
DuckDB
Timed Linux Kernel Compilation
OpenVINO
Tachyon
OpenVINO:
  Road Segmentation ADAS FP16 - CPU
  Vehicle Detection FP16 - CPU
ASKAP:
  tConvolve MPI - Gridding
  tConvolve MPI - Degridding
oneDNN
OpenVINO:
  Handwritten English Recognition FP16-INT8 - CPU
  Person Detection FP16 - CPU
  Person Detection FP32 - CPU
  Handwritten English Recognition FP16 - CPU
  Face Detection Retail FP16 - CPU
  Face Detection FP16 - CPU
ASTC Encoder
OpenVINO
Cpuminer-Opt
OpenVINO
NAS Parallel Benchmarks
ASTC Encoder
Timed Godot Game Engine Compilation
SVT-AV1
RawTherapee
Timed Gem5 Compilation
SVT-AV1
OpenVINO
Timed Mesa Compilation
SVT-AV1:
  Preset 13 - Bosphorus 4K
  Preset 12 - Bosphorus 4K
PostgreSQL:
  100 - 1000 - Read Write
  100 - 1000 - Read Write - Average Latency
DuckDB
CPU Power Consumption Monitor:
  Phoronix Test Suite System Monitoring
  CPU Power Consumption Monitor
  Matrix 3D Math
Stress-NG
Stockfish:
  CPU Power Consumption Monitor
  Total Time
Stockfish
LuxCoreRender:
  CPU Power Consumption Monitor
  DLSC - CPU
LuxCoreRender
John The Ripper:
  CPU Power Consumption Monitor
  MD5
John The Ripper