AMD EPYC 9554 EPYC 9654 Benchmarks Genoa

Benchmarks for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2211102-PTS-GENOAEXT88
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

Limit displaying results to tests within:

Timed Code Compilation 4 Tests
C/C++ Compiler Tests 4 Tests
CPU Massive 8 Tests
Creator Workloads 8 Tests
Cryptocurrency Benchmarks, CPU Mining Tests 2 Tests
Cryptography 3 Tests
Database Test Suite 2 Tests
Encoding 4 Tests
Game Development 2 Tests
HPC - High Performance Computing 5 Tests
Imaging 3 Tests
Machine Learning 3 Tests
Multi-Core 11 Tests
OpenMPI Tests 2 Tests
Programmer / Developer System Benchmarks 4 Tests
Python Tests 7 Tests
Server 2 Tests
Server CPU Tests 7 Tests
Video Encoding 4 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

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

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
EPYC 9654 2P
October 15 2022
  3 Hours, 46 Minutes
EPYC 9654 2P Repeat
October 15 2022
  2 Hours, 13 Minutes
EPYC 9654 2P - Perf M
October 15 2022
  2 Hours, 16 Minutes
EPYC 9654
October 15 2022
  2 Hours, 19 Minutes
EPYC 9654 Repeat
October 15 2022
  2 Hours, 20 Minutes
EPYC 9654 Power
October 16 2022
  2 Hours, 15 Minutes
9654 Power
October 17 2022
  2 Hours, 5 Minutes
AMD EPYC 75F3 32-Core
October 29 2022
  11 Hours, 39 Minutes
75F32
October 30 2022
  11 Hours, 50 Minutes
9554 TwoP
November 01 2022
  4 Hours, 1 Minute
EPYC 9554
November 03 2022
  1 Hour, 15 Minutes
8380 2p
November 05 2022
  6 Hours, 45 Minutes
Invert Hiding All Results Option
  4 Hours, 24 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):

EPYC 9654 2P

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

EPYC 9654 2P Repeat

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

EPYC 9654 2P - Perf M

Processor: 2 x AMD EPYC 9654 96-Core @ 3.71GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080

EPYC 9654

EPYC 9654 Repeat

EPYC 9654 Power

9654 Power

Changed Processor to AMD EPYC 9654 96-Core @ 3.71GHz (96 Cores / 192 Threads).

Changed Memory to 768GB.

AMD EPYC 75F3 32-Core

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

75F32

Processor: AMD EPYC 75F3 32-Core @ 2.95GHz (32 Cores / 64 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 256GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710

9554 TwoP

Processor: 2 x AMD EPYC 9554 64-Core @ 3.76GHz (128 Cores / 256 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

EPYC 9554

Changed Processor to AMD EPYC 9554 64-Core @ 3.76GHz (64 Cores / 128 Threads).

Changed Memory to 768GB.

8380 2p

Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Ice Lake IEH, Memory: 512GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0xd000375
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: 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_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

AMD EPYC 9554 EPYC 9654 Benchmarks GenoaProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerVulkanCompilerFile-SystemScreen ResolutionEPYC 9654 2PEPYC 9654 2P RepeatEPYC 9654 2P - Perf MEPYC 9654EPYC 9654 RepeatEPYC 9654 Power9654 PowerAMD EPYC 75F3 32-Core75F329554 TwoPEPYC 95548380 2p2 x AMD EPYC 9654 96-Core @ 3.71GHz (192 Cores / 384 Threads)AMD Titanite_4G (RTI1002E BIOS)AMD Device 14a41520GB3841GB Micron_9300_MTFDHAL3T8TDPASPEEDVGA HDMIBroadcom NetXtreme BCM5720 PCIeUbuntu 22.106.0.0-060000rc3daily20220904-generic (x86_64)GNOME ShellX Server 1.21.1.31.3.211GCC 12.2.0ext41920x1080AMD EPYC 9654 96-Core @ 3.71GHz (96 Cores / 192 Threads)768GBAMD EPYC 75F3 32-Core @ 2.95GHz (32 Cores / 64 Threads)AMD DAYTONA_X (RYM1009B BIOS)AMD Starship/Matisse256GBVE2282 x Mellanox MT277102 x AMD EPYC 9554 64-Core @ 3.76GHz (128 Cores / 256 Threads)AMD Titanite_4G (RTI1002E BIOS)AMD Device 14a41520GBVGA HDMIBroadcom NetXtreme BCM5720 PCIeAMD EPYC 9554 64-Core @ 3.76GHz (64 Cores / 128 Threads)768GB2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads)Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS)Intel Ice Lake IEH512GBVE2282 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFPOpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- EPYC 9654 2P: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- EPYC 9654 2P Repeat: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- EPYC 9654 2P - Perf M: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- EPYC 9654: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- EPYC 9654 Repeat: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- EPYC 9654 Power: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- 9654 Power: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- AMD EPYC 75F3 32-Core: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173- 75F32: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173- 9554 TwoP: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- EPYC 9554: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d- 8380 2p: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0xd000375Java Details- EPYC 9654 2P, AMD EPYC 75F3 32-Core, 75F32, 9554 TwoP, EPYC 9554, 8380 2p: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python Details- Python 3.10.6Security Details- EPYC 9654 2P: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- EPYC 9654 2P Repeat: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- EPYC 9654 2P - Perf M: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- EPYC 9654: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- EPYC 9654 Repeat: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- EPYC 9654 Power: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 9654 Power: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- AMD EPYC 75F3 32-Core: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 75F32: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 9554 TwoP: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- EPYC 9554: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected- 8380 2p: 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_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

EPYC 9654 2PEPYC 9654 2P RepeatEPYC 9654 2P - Perf MEPYC 9654EPYC 9654 RepeatEPYC 9654 Power9654 PowerAMD EPYC 75F3 32-Core75F329554 TwoPEPYC 95548380 2pLogarithmic Result OverviewPhoronix Test SuiteTensorFlow7-Zip CompressionTimed Linux Kernel CompilationTimed Node.js CompilationY-CruncherOpenFOAMx264Timed Godot Game Engine CompilationOpenRadiosslibavif avifencx265JPEG XL Decoding libjxlTimed Mesa CompilationJPEG XL libjxl

AMD EPYC 9554 EPYC 9654 Benchmarks Genoacpuminer-opt: LBC, LBRY Creditstensorflow: CPU - 256 - AlexNetstress-ng: Socket Activitydeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamtensorflow: CPU - 32 - VGG-16tensorflow: CPU - 512 - GoogLeNetdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamstress-ng: MMAPcpuminer-opt: Ringcointensorflow: CPU - 512 - ResNet-50tensorflow: CPU - 256 - GoogLeNettensorflow: CPU - 16 - VGG-16deepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamcpuminer-opt: scrypttensorflow: CPU - 256 - ResNet-50tensorflow: CPU - 64 - AlexNetdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamopenfoam: drivaerFastback, Medium Mesh Size - Execution Timestress-ng: CPU Cachecpuminer-opt: Quad SHA-256, Pyriteblender: Classroom - CPU-Onlycpuminer-opt: Blake-2 Sblender: Pabellon Barcelona - CPU-Onlyblender: Barbershop - CPU-Onlycpuminer-opt: Skeincoinblender: BMW27 - CPU-Onlycompress-7zip: Decompression Ratingtensorflow: CPU - 32 - AlexNetstress-ng: Glibc Qsort Data Sortingstress-ng: Glibc C String Functionsblender: Fishy Cat - CPU-Onlytensorflow: CPU - 64 - ResNet-50tensorflow: CPU - 64 - GoogLeNetstress-ng: Matrix Mathtensorflow: CPU - 16 - AlexNetcpuminer-opt: x25xstress-ng: Cryptocpuminer-opt: Magistress-ng: CPU Stressstress-ng: Mutexstress-ng: Mallocstress-ng: Vector Mathstress-ng: SENDFILEdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamstress-ng: Forkingcpuminer-opt: Triple SHA-256, Onecoinbuild-linux-kernel: allmodconfigtensorflow: CPU - 32 - ResNet-50cpuminer-opt: Deepcoinhammerdb-mariadb: 64 - 500hammerdb-mariadb: 64 - 500hammerdb-mariadb: 64 - 250xmrig: Wownero - 1Mhammerdb-mariadb: 64 - 250stress-ng: Context Switchingstress-ng: MEMFDtensorflow: CPU - 32 - GoogLeNetcompress-7zip: Compression Ratingxmrig: Monero - 1Mstress-ng: Memory Copyingtensorflow: CPU - 16 - ResNet-50stress-ng: Semaphorestensorflow: CPU - 16 - GoogLeNetopenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenradioss: Rubber O-Ring Seal Installationdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamopenradioss: INIVOL and Fluid Structure Interaction Drop Containerstress-ng: Futexbuild-nodejs: Time To Compiledeepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Streamdeepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Streamopenradioss: Cell Phone Drop Testdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamy-cruncher: 1By-cruncher: 500Mx264: Bosphorus 4Kbuild-linux-kernel: defconfigspacy: en_core_web_trfdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamcpuminer-opt: Garlicointensorflow: CPU - 512 - VGG-16tensorflow: CPU - 256 - VGG-16cpuminer-opt: Myriad-Groestltensorflow: CPU - 512 - AlexNettensorflow: CPU - 64 - VGG-16openradioss: Bird Strike on Windshielddeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamhbase: 10000 - Rand Read - 256build-godot: Time To Compileopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timesmhasher: t1ha0_aes_avx2 x86_64openfoam: drivaerFastback, Small Mesh Size - Mesh Timestress-ng: Atomicdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamhbase: 10000 - Increment - 1avifenc: 6avifenc: 6, Losslesshbase: 10000 - Increment - 1hbase: 10000 - Rand Read - 256deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamavifenc: 10, Losslessjpegxl: PNG - 100hbase: 1000000 - Rand Read - 1openfoam: motorBike - Execution Timehbase: 1000000 - Rand Read - 1x265: Bosphorus 4Kjpegxl-decode: Allhbase: 10000 - Rand Write - 256deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamstress-ng: System V Message Passingjpegxl: JPEG - 100x264: Bosphorus 1080pjpegxl-decode: 1x265: Bosphorus 1080pstress-ng: NUMAdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streambuild-mesa: Time To Compilehbase: 10000 - Rand Read - 1avifenc: 2hbase: 10000 - Rand Read - 1hbase: 10000 - Increment - 256openradioss: Bumper Beamffmpeg: libx265 - Platformffmpeg: libx265 - Platformjpegxl: JPEG - 90hbase: 1000000 - Increment - 1smhasher: t1ha2_atonceavifenc: 0jpegxl: PNG - 80jpegxl: JPEG - 80hbase: 1000000 - Increment - 1jpegxl: PNG - 90ffmpeg: libx265 - Video On Demandffmpeg: libx265 - Video On Demandffmpeg: libx265 - Uploadffmpeg: libx265 - Uploadsmhasher: wyhashsmhasher: fasthash32smhasher: SHA3-256smhasher: MeowHash x86_64 AES-NIsmhasher: Spooky32smhasher: FarmHash128smhasher: FarmHash32 x86_64 AVXhbase: 10000 - Rand Read - 500hammerdb-mariadb: 64 - 100hammerdb-mariadb: 64 - 100spacy: en_core_web_lghbase: 10000 - Rand Read - 500ffmpeg: libx264 - Platformffmpeg: libx264 - Video On Demandffmpeg: libx264 - Platformffmpeg: libx264 - Video On Demandffmpeg: libx264 - Liveffmpeg: libx264 - Liveffmpeg: libx264 - Uploadffmpeg: libx264 - Uploadhbase: 10000 - Increment - 256hbase: 10000 - Increment - 500ffmpeg: libx265 - Liveffmpeg: libx265 - Liveclickhouse: 100M Rows Web Analytics Dataset, Second Runhbase: 10000 - Rand Write - 1clickhouse: 100M Rows Web Analytics Dataset, Third Runhbase: 10000 - Rand Write - 256hbase: 10000 - Rand Write - 1stress-ng: IO_uringclickhouse: 100M Rows Web Analytics Dataset, First Run / Cold Cachehbase: 1000000 - Increment - 256hbase: 1000000 - Increment - 256hbase: 10000 - Increment - 500hbase: 1000000 - Increment - 500hbase: 1000000 - Increment - 500hbase: 1000000 - Scan - 256hbase: 1000000 - Scan - 1hbase: 10000 - Scan - 1smhasher: MeowHash x86_64 AES-NIsmhasher: t1ha0_aes_avx2 x86_64smhasher: FarmHash32 x86_64 AVXsmhasher: t1ha2_atoncesmhasher: FarmHash128smhasher: fasthash32smhasher: Spooky32smhasher: SHA3-256smhasher: wyhashEPYC 9654 2PEPYC 9654 2P RepeatEPYC 9654 2P - Perf MEPYC 9654EPYC 9654 RepeatEPYC 9654 Power9654 PowerAMD EPYC 75F3 32-Core75F329554 TwoPEPYC 95548380 2p1279.03836.988660.45441.712227.4198147.32322.2341967.2269111.54475.541318.3224663.579107.8582617.6522.8872.867.531323090296.2910.5657.85157.71153.2998.040897.4788100.4139.57108.5694057421.5162.4419.919399124.5175.695.54116.8710.1544.925104.7719.7252515122.13111.051672.8981.58208.98114.3987964.8329974.015327033.863133.96745102425.2724.84760243.01121812.4225.3235319388592144.04594.3220.8551792522.94275.9772.63230.7348.5399.026520.08914234.654669886476.49.236920135.263.5629.579.29143459.4831218.058803.42216.0154190.5420577.1521167.1739795.994681348133948215524615993625907199312710514942945.1120.47226.38720.09747.82822.07626.8551829.79813.9881326.14840.23459.82443.932244.1857147.82327.4740.06982.4985110.01545.591320.4156671.4008104.2302717.7523.2872.787.471309354286.4510.4659.49166.76161.4197.285697.6967100.52540.61108.9493094622.4862.6919.631093126.019.866101.302176.1595.4255.0334198.499117.0627.290236.63327.213336.736410.134.947104.5619.65824615.1748193.0775123.2113.241669.8779.43208.25113.97725.2274191.206811.345688.0745969.4243966.601533.335118.67648101829.925.53952742.62582.4155.21142.57914.2880.795.9805224.3275.7572.5440.7265.9648.5172.7697.501820.13134.52675.769.0519789.9663.1089.499.329.4731247.368803.35216.0254192.2220580.2821167.4839794.951346745.1120.47226.38720.09747.82522.07727.6171829.04313.9681198.78763.182755.34437.651949.0047145.44320.9940.86858.8745108.07535.771195.8283616.4604112.6945120.1125.6781.918.471176187289.6511.5158.76168.86160.484.479884.2018108.58439.86108.7190831222.2963.0523.083579131.1110.030699.637282.5298.3025.2709189.530818.6429.407333.994928.974534.503610.1644.98365.3720.48625855.4359183.7787110.75104.391587.0575.72216.62125.40965.2827189.206211.750285.0441132.37751131.689133.419141.1001102396.4627.07822149.14092.4225.241155.29564.2440.85.9018122.86284.1980.09480.73200.6549.0270.5111.482120.0135.07482.839.2820135.1463.0789.689.299.6231248.918803.4216.1554144.8520576.7721167.2939795.431341545.11420.47226.38720.09747.82922.07826.6621829.20714.2151258.48402.57457.61530.051010.1033165473.451.31449.3802146.3834.42625.932318.5908246.878438.6748.75143.4215.69652935581.1720.3899.93295.82360.1843.005142.2142173.49476.17227.8867734252.44130.4729.73861568.049.8478101.5028103.31122.9115.2151191.606122.9928.10435.574828.019335.68186.5163.96965.824.51932345.2236191.324469.5766.421373.0261.53120.27119.05255.2064191.995311.465187.17751115.49631110.003733.124119.71925102376.0323.91939947.47462.295.161150.60574.1240.824.30131.16317.8376.64330.77211.749.7891.88106.551820.7533.7573.399.5720134.6962.2999.959.689.931198.528803.52216.0854220.2320581.5521167.6439796.131355145.1120.47226.38720.09747.82522.07726.7031829.74913.9681257.65400.729357.45529.621008.1787164.66471.9251.6444.9684146.45838.94624.2323316.9094246.8270238.7149.15143.5615.43653127580.9720.83100.91310.69358.8242.988443.1327174.42675.61225.0667319852.44144.3828.79171566.599.8059101.9321104.23122.7495.212191.721722.7328.215835.434128.092835.58866.543.959109.7424.6333075.2613189.925869.5466.491369.5461.54121.04119.53595.1744193.185611.406287.62561115.90131112.484633.399131.63939102315.3726.38931247.53772.3215.226151.37054.4270.824.3512231.32316.4776.84880.73265.7749.8391.8107.66820.82833.83577.939.5820135.1962.2749.949.689.8731201.28775.9211.8254212.9620582.5721166.2139795.041351745.11220.47226.38720.09747.82522.07726.6631871.00413.9681331.08424.433560.63558.921127.3599172.75495.4453.71497.6848150.26870.59662.79337.0033237.4959733.6343.61127.4514737688599.9818.21100.8316.07366.7348.840949.0518157.73477.29214.9871764753.4145.1124.50917361.499.77102.307490.53117.065.0333198.520820.4526.669737.487626.806337.29686.473.87265.523.19531865.1375194.499473.6570.931467.7165.62113.86112.88845.2442190.616411.206489.1886971.9156972.479732.438113.75167102400.9523.77435742.52172.3115.083142.10024.0780.824.35231.57318.972.3310.73221.2549.7193.5496.226420.39833.770.829.6220135.1761.7319.959.79.931197.928802.19216.1354202.2720582.7921168.0139795.021352745.11220.47726.38720.09747.82522.0826.6831830.33613.9681332.68426.363361.12557.521131.4819172.84495.3553.84499.0893149.71866.54666.0508339.5245240.3849533.5543.73127.7813.73739434600.3118.14100.73305.7367.1448.856648.9538157.45576.4922971957952.97144.1525.75680961.5190.38117.30920.256.3543.91766.623.248331473.8670.841461.665.84111.68112.4239970.5421972.119232.76111.2760410239425.70776842.36452.2975.227141.03424.0990.8131.33320.1672.01020.7450.795.968620.49633.77169.939.6420135.162.03710.059.779.9831245.478803.73216.0854175.0820577.0121167.1139794.111352645.10620.47226.38720.09747.82422.07927.6171829.02813.98489790148.1252568.4498.97387.7666.93268.728508.453864.5622.8165.837.41136.3518554.9922.08129.35201.3251102.8223662.5032588.14285840106.781493100131.98417.1228734041.47247362113.41434.575593873.252.1520.6464.47141942.9592.211399.5955423.941432.9885344.0418586817.03140011498.66202508.2645248.6824.167324.226980179.88645080390.84620.113177025865.214657098.151076.6865.2828057224698.16643.9518.536533513.1657.6756.4490168.2325.478139.2348197.462573853.71230.15311.71685.31738.5361.328816.303761.439416.274414.5057.15159.0141.343189410.46795.48444386.98.398.325323.29155.167.99143.06161.22696.9821143.165619.739150.6446658.3907658.974537752.216140.0479727.369771160435.9859.47342373.2656.5364097298877155.25894.7250.75544.667411821628.64332.7336779679.34718704140.320.69205.0547.484.07926.64117.10526.15118640.949518998191.019.517477.92910.029.59133419.86529136543068691418251821402.4134387.0219532695467389.93376.512062123070154170435211398389690148.2852321.4798.94577.866.56269.3961511.983795.4322.6165.337.4136.5358554.8821.95129.43200.8362102.1985664.54696100.49286410106.731476790130.92418.9128734041.41247381112.33432.85591675.7752.2320.5265.21141780.3193.371398.4755510.141433.685478.8818679911.5139310693.59202547.6648949.2824.097324.116580711.89644780393.15319.973176025841.114724434.091078.6965.5327756824506.86634.2418.56558615.4358.2756.2467768.4725.592539.0595197.962708005.03230.01511.704285.402138.9661.428916.277461.258616.322414.4627.11873.9741.401187310.557494.67422988.418.398.3369090154.558141.83161.35266.9796143.216419.84450.378660.7807660.989544552.07138.8770928.369974165666.6559.33012673.3046.6513631266084156.19344.9450.75764.666761310828.73320.1650787779.538711498.670.71217.7546.784.18925.07117.018826.31918941.0245071113992.19.339077.8489.779.45110719.68589136442789411299411702401.9934383.6718712702767158.17377.332026125256153938440911252758887584131234237538546301366.7179908.39629.905365.82494.21608.98914187.21591.17163.93378.9650.73709.16533939.46127.49645.23997.8874511.6331129.8374933.95179580024.2865989032.0497.1216318309.93945724387.922210.092808127113.669.01194.59683986.91219.456438.59250785.926463.85366719.7777897090.19576947450.43832532.652643264.3370.210768.568923702560590113.86246.7512279026791115781232297285.7287077907313.682624.53138.4477644782027.720141.2429.8319242578.6987.3820.90601879.6810.859892.002487.863307879.3799.5235.3104188.136117.2738.211626.16538.338326.07867.9964.38997.720.12221186.002166.39355824097.3292.729939.681649.6379.07165.5101.42583.921254.895211.487.6574911.1053909.124332.41116.4884980341.3622.631593241171.3839.70452.4045.316124.80894.250.825.9227423.42278.0664.0098962692.120.69262.7948.8869.28904.290.085120.19534.9472.64133.93132242156.569.2915798.7662.2629.729.539.64135.46128618955.9290.8773505427.7824475.936903.87169.5542531.5616145.6916608.9231226.87371098623113540154.7092761249.1148.96154.24541651522.885944059220.66198.92808819812.69127.9239.47773307865946.5557.7426.25833.66126.8161.34528.34934.2642331.75318.1951.5147.62298.96059527376426.56186.8756565029.84870149.82117.93133.45923.196.363.358.7925.37754.32108.8134.607103.9422580231.0423.8130942.3425.2474.1060.8131.73341.130.7550.4820.72133.47570.489.7115799.0961.17110.089.7910.0124509.446907.82169.5742535.2316149.316609.2931226.5757.73226.25833.65625.75961.35128.32934.2682331.57518.085416520654.988958.77215.809432.24298.78826.80573557.26473783.01291.0828.64331.53392222.7780.79498.31452.3605220.229365.3070415.9690947061.97438431077.12244.0262220023.57363204402.14900.619412669.5131.1175.88249.3320154.03306.582657.9985416.532250.31213661.3343926099.15210690160.09304766.81958441.0447.748647.652564952.621323260228.52967.33632701027994426746470409871079384007912.583857.49209.540397826231.610739.6354.5112534447.06167.0542.11775182.1512.839477.83146.631294371.34170.9928.3167120.070933.5951.476619.423250.777419.690910.4995.32451.830.55638689.9106100.82293139037.0435.7842780701.2733.85144.33185.27934.4386224.846919.13952.2285837.5412836.902944.984165.0807864712.6734.432062197789.2348.34113.3657.447181.35625.8060.5824.77246.9188.30126498292.380.56193.6536.9474.49693.4120.345626.70244.40786.3175.80415145843.097.4220601.6179.6677.747.527.74174.6643.37116.4121.6927859.548453.99208.3145558.1817883.7119206.2532568.144669010849411140187.3440.5640.44186.7427.68182.45239.5310.54114.0644.27414.29398.0364705.57382.2951.41925.68731.90925.89750.21828.83137.6991876.36319.627OpenBenchmarking.org

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 3.20.3Algorithm: LBC, LBRY Credits75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core200K400K600K800K1000K89690416520854630897901. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: AlexNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat30060090012001500148.28654.981366.711332.68148.121258.481279.031198.781326.141331.081257.65

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Socket Activity75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core20K40K60K80K100K52321.478958.7779908.3952568.441. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat200400600800100098.95215.81629.91426.3698.97402.57836.99763.18840.23424.43400.73

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: VGG-1675F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat15304560757.8032.2465.8261.127.7651.5157.6160.4555.3459.8260.6357.45

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: GoogLeNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat12024036048060066.56298.78494.20557.5266.93530.05441.71437.65443.93558.92529.62

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat5001000150020002500269.40826.811608.991131.48268.731010.102227.421949.002244.191127.361008.18

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MMAP75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core9001800270036004500511.983557.264187.21508.451. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

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 3.20.3Algorithm: Ringcoin75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core100020003000400050003795.434737.00591.173864.561. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: ResNet-5075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat408012016020022.6183.01163.93172.8422.81165.00147.32145.44147.82172.75164.66

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: GoogLeNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat11022033044055065.33291.08378.96495.3565.83473.40322.23320.99327.47495.44471.92

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: VGG-1675F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat12243648607.4028.6450.7353.847.4147.6251.3141.0040.8640.0653.7151.60

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2004006008001000136.54331.53709.17499.09136.35449.38967.23858.87982.50497.68444.97

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 3.20.3Algorithm: scrypt75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core8001600240032004000554.882222.773939.46554.991. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: ResNet-5075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat30609012015021.9580.79127.49149.7122.08146.30111.54108.07110.01150.26146.45

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: AlexNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2004006008001000129.43498.31645.23866.54129.35834.42475.54535.77545.59870.59838.94

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat30060090012001500200.84452.36997.89666.05201.33625.931318.321195.831320.42662.79624.23

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat140280420560700102.20220.23511.63339.52102.82318.59663.58616.46671.40337.00316.91

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Execution Time75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat140280420560700664.55365.31129.84240.38662.50298.96246.88107.86112.69104.23237.50246.83-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfoamToVTK -ldynamicMesh -lfileFormats-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -llagrangian -lgenericPatchFields -lOpenFOAM -ldl -lm

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU Cache75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core20406080100100.4915.9633.9588.141. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

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 3.20.3Algorithm: Quad SHA-256, Pyrite75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core400K800K1200K1600K2000K28641090947017958002858401. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Classroom - Compute: CPU-Only75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat20406080100106.7361.9724.2033.55106.7838.6717.6520.1117.7533.6338.71

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 3.20.3Algorithm: Blake-2 S75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core2M4M6M8M10M14767904384310865989014931001. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Pabellon Barcelona - Compute: CPU-Only75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat306090120150130.9277.1232.0443.73131.9848.7522.8825.6723.2843.6149.15

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Barbershop - Compute: CPU-Only75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat90180270360450418.91244.0297.12127.78417.12143.4272.8681.9172.78127.45143.56

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 3.20.3Algorithm: Skeincoin75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core300K600K900K1200K1500K28734062220016318302873401. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: BMW27 - Compute: CPU-Only75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat91827364541.4123.579.9313.7341.4715.697.538.477.4714.0015.43

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 Rating75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat300K600K900K1200K1500K2473813632049457247394342473625273766529351323090117618713093547376886531271. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: AlexNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat130260390520650112.33402.14387.92600.31113.41581.17296.29289.65286.45599.98580.97

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc Qsort Data Sorting75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core5001000150020002500432.80900.612210.09434.571. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc C String Functions75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core6M12M18M24M30M5591675.779412669.5128081271.005593873.201. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Fishy Cat - Compute: CPU-Only75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat122436486052.2331.1113.6018.1452.1520.3810.5611.5110.4618.2120.83

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: ResNet-5075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010020.5275.8869.01100.7320.6499.9357.8558.7659.49100.80100.91

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: GoogLeNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat7014021028035065.21249.30194.59305.7064.47295.82157.71168.86166.76316.07310.69

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Matrix Math75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core150K300K450K600K750K141780.31320154.03683986.91141942.951. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: AlexNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat9018027036045093.37306.58219.45367.1492.21426.56360.18153.29160.40161.41366.73358.82

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 3.20.3Algorithm: x25x75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core140028004200560070001398.472657.996438.591399.591. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Crypto75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core50K100K150K200K250K55510.1485416.53250785.9255423.941. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

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 3.20.3Algorithm: Magi75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core140028004200560070001433.602250.316463.851432.981. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU Stress75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core80K160K240K320K400K85478.88213661.33366719.7785344.041. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Mutex75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core20M40M60M80M100M18679911.5043926099.1577897090.1918586817.031. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Malloc75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core120M240M360M480M600M139310693.59210690160.09576947450.43140011498.661. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Vector Math75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core200K400K600K800K1000K202547.60304766.81832532.65202508.201. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: SENDFILE75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core600K1200K1800K2400K3000K648949.28958441.042643264.33645248.681. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010024.1047.7570.2148.8624.1743.0198.0484.4897.2948.8442.99

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010024.1247.6568.5748.9524.2342.2197.4884.2097.7049.0543.13

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Forking75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core20K40K60K80K100K80711.8964952.622370.0080179.881. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

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 3.20.3Algorithm: Triple SHA-256, Onecoin75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core500K1000K1500K2000K2500K644780132326025605906450801. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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 5.18Build: allmodconfig75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat90180270360450393.15228.53113.86157.46390.85186.87173.49100.41108.58100.53157.73174.43

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: ResNet-5075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010019.9767.3346.7576.4920.1176.1739.5739.8640.6177.2975.61

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 3.20.3Algorithm: Deepcoin75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core30K60K90K120K150K3176063270122790317701. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

HammerDB - MariaDB

This is a MariaDB MySQL database server benchmark making use of the HammerDB benchmarking / load testing tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Minute, More Is BetterHammerDB - MariaDB 10.9.3Virtual Users: 64 - Warehouses: 5008380 2p9554 TwoP20K40K60K80K100K102799267911. (CXX) g++ options: -fPIC -pie -fstack-protector -O2 -shared -lrt -lpthread -lz -ldl -lm -lstdc++

OpenBenchmarking.orgNew Orders Per Minute, More Is BetterHammerDB - MariaDB 10.9.3Virtual Users: 64 - Warehouses: 5008380 2p9554 TwoP9K18K27K36K45K44267115781. (CXX) g++ options: -fPIC -pie -fstack-protector -O2 -shared -lrt -lpthread -lz -ldl -lm -lstdc++

OpenBenchmarking.orgNew Orders Per Minute, More Is BetterHammerDB - MariaDB 10.9.3Virtual Users: 64 - Warehouses: 2508380 2p9554 TwoP10K20K30K40K50K46470123221. (CXX) g++ options: -fPIC -pie -fstack-protector -O2 -shared -lrt -lpthread -lz -ldl -lm -lstdc++

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: 1M75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core20K40K60K80K100K25841.140987.097285.725865.21. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

HammerDB - MariaDB

This is a MariaDB MySQL database server benchmark making use of the HammerDB benchmarking / load testing tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Minute, More Is BetterHammerDB - MariaDB 10.9.3Virtual Users: 64 - Warehouses: 2508380 2p9554 TwoP20K40K60K80K100K107938287071. (CXX) g++ options: -fPIC -pie -fstack-protector -O2 -shared -lrt -lpthread -lz -ldl -lm -lstdc++

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Context Switching75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core3M6M9M12M15M14724434.094007912.587907313.6814657098.151. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MEMFD75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core80016002400320040001078.693857.492624.531076.681. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: GoogLeNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat5010015020025065.53209.50138.44229.0065.28227.88108.56108.71108.94214.98225.06

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 Rating75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat200K400K600K800K1000K2775684039787764477195792805725656506773429405749083129309467176476731981. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

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: 1M75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core20K40K60K80K100K24506.826231.682027.724698.11. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Memory Copying75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core4K8K12K16K20K6634.2410739.6320141.246643.951. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: ResNet-5075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat122436486018.5054.5129.8352.9718.5352.4421.5122.2922.4853.4052.44

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Semaphores75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core4M8M12M16M20M6558615.4312534447.0619242578.696533513.161. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: GoogLeNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat408012016020058.27167.0587.38144.1557.67130.4762.4463.0562.69145.11144.38

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution Time75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat132639526556.2542.1220.9125.7656.4529.8529.7419.9223.0819.6324.5128.79-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfoamToVTK -ldynamicMesh -lfileFormats-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -llagrangian -lgenericPatchFields -lOpenFOAM -ldl -lm

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/. 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 2022.10.13Model: Rubber O-Ring Seal Installation75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat30609012015068.4782.1579.6861.5168.2349.8268.04124.51131.11126.0161.4966.59

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat61218243025.592512.839410.859825.47819.847810.03069.86609.77009.8059

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010039.0677.8392.0039.23101.5099.64101.30102.31101.93

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/. 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 2022.10.13Model: INIVOL and Fluid Structure Interaction Drop Container75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat4080120160200197.96146.6387.8690.38197.46117.93103.3175.6082.5276.1590.53104.23

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Futex75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core700K1400K2100K2800K3500K2708005.031294371.343307879.372573853.711. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

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 18.8Time To Compile75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat50100150200250230.02170.9999.52117.31230.15133.46122.9195.5498.3095.43117.06122.75

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat369121511.70428.31675.310411.71605.21515.27095.03345.03335.2120

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat408012016020085.40120.07188.1485.32191.61189.53198.50198.52191.72

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/. 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 2022.10.13Model: Cell Phone Drop Test75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat91827364538.9633.5917.2720.2538.5323.1922.9916.8718.6417.0620.4522.73

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat142842567061.4351.4838.2161.3328.1029.4127.2926.6728.22

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat91827364516.2819.4226.1716.3035.5733.9936.6337.4935.43

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat142842567061.2650.7838.3461.4428.0228.9727.2126.8128.09

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat91827364516.3219.6926.0816.2735.6834.5036.7437.3035.59

Y-Cruncher

Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.10.9513Pi Digits To Calculate: 1B75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat4812162014.46210.4997.9966.35414.5056.3606.51610.15410.16410.1306.4706.540

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.10.9513Pi Digits To Calculate: 500M75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2468107.1185.3244.3893.9177.1513.3003.9694.9254.9834.9473.8723.959

x264

This is a multi-threaded test of the x264 video encoder run on the CPU with a choice of 1080p or 4K video input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2022-02-22Video Input: Bosphorus 4K75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010073.9751.8097.7066.6059.0158.7965.80104.7765.37104.5665.50109.741. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -flto

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 5.18Build: defconfig75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat91827364541.4030.5620.1223.2541.3425.3824.5219.7320.4919.6623.2024.63

spaCy

The spaCy library is an open-source solution for advanced neural language processing (NLP). The spaCy library leverages Python and is a leading neural language processing solution. This test profile times the spaCy CPU performance with various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgtokens/sec, More Is BetterspaCy 3.4.1Model: en_core_web_trf75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat800160024003200400018733868211833141894323425152585246131863307

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat369121510.55749.91066.002010.46705.22365.43595.17485.13755.2613

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat408012016020094.67100.82166.3995.48191.32183.78193.08194.50189.93

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 3.20.3Algorithm: Garlicoin75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core12K24K36K48K60K2988.4131390.0058240.004386.901. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: VGG-1675F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat3060901201508.3937.0497.3273.868.3969.57122.13110.75123.2073.6569.54

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: VGG-1675F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat3060901201508.3335.7892.7270.848.3266.42111.05104.39113.2470.9366.49

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 3.20.3Algorithm: Myriad-Groestl75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core15K30K45K60K75K69090.0042780.009939.685323.291. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: AlexNet75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat400800120016002000154.55701.271649.631461.60155.161373.021672.891587.051669.871467.711369.54

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: VGG-1675F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat204060801008.0033.8579.0765.847.9954.3261.5381.5875.7279.4365.6261.54

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/. 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 2022.10.13Model: Bird Strike on Windshield75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat50100150200250141.83144.33165.50111.68143.06108.81120.27208.98216.62208.25113.86121.04

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat4080120160200161.35185.28101.43112.42161.23119.05114.40125.41113.98112.89119.54

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2468106.97964.43863.92106.98215.20645.28275.22745.24425.1744

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat60120180240300143.22224.85254.90143.17192.00189.21191.21190.62193.19

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat51015202519.8419.1411.4019.7411.4711.7511.3511.2111.41

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010050.3852.2387.6650.6487.1885.0488.0789.1987.63

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2004006008001000660.78837.54911.11970.54658.391115.50964.831132.38969.42971.921115.90

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2004006008001000660.99836.90909.12972.12658.971110.00974.021131.69966.60972.481112.48

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Read - Clients: 25675F32AMD EPYC 75F3 32-CoreEPYC 9654 2P100200300400500445377270

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 3.2.3Time To Compile75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat122436486052.0744.9832.4132.7652.2234.6133.1233.8633.4233.3432.4433.40

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh Time75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat4080120160200138.88165.08116.49111.28140.05103.94119.72133.97141.10118.68113.75131.64-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfoamToVTK -ldynamicMesh -lfileFormats-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -llagrangian -lgenericPatchFields -lOpenFOAM -ldl -lm

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha0_aes_avx2 x86_648380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat20K40K60K80K100K64712.6780341.36102394.0080231.04102376.03102425.27102396.46101829.90102400.95102315.371. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: t1ha0_aes_avx2 x86_648380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat61218243025.6926.2620.4726.2620.4720.4720.4720.4720.4820.471. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: t1ha0_aes_avx2 x86_64

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh Time75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat81624324028.3734.4322.6325.7127.3723.8123.9224.8527.0825.5423.7726.39-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfoamToVTK -ldynamicMesh -lfileFormats-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels-lfiniteVolume -lmeshTools -lparallel -lregionModels1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -llagrangian -lgenericPatchFields -lOpenFOAM -ldl -lm

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Atomic75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core50K100K150K200K250K165666.65197789.23241171.38160435.981. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat132639526559.3348.3439.7042.3659.4747.4743.0149.1442.6342.5247.54

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 10000 - Test: Increment - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P60120180240300267237181

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 675F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat0.75711.51422.27133.02843.78553.3043.3652.4042.2973.2652.3422.2902.4222.4222.4152.3112.3211. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6, Lossless75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2468106.6517.4475.3165.2276.5365.2475.1615.3235.2415.2105.0835.2261. (CXX) g++ options: -O3 -fPIC -lm

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Increment - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P11002200330044005500363140975319

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Read - Clients: 25675F32AMD EPYC 75F3 32-CoreEPYC 9654 2P80K160K240K320K400K266084298877388592

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat4080120160200156.19181.36124.81141.03155.26150.61144.05155.30142.58142.10151.37

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 10, Lossless75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat1.30642.61283.91925.22566.5324.9455.8064.2504.0994.7254.1064.1244.3224.2444.2884.0784.4271. (CXX) g++ options: -O3 -fPIC -lm

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 10075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat0.18450.3690.55350.7380.92250.750.580.820.810.750.810.820.800.800.790.820.821. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Random Read - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P20406080100765455

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: motorBike - Execution Time75F329554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat1.34562.69124.03685.38246.7284.666765.922744.667414.301005.901815.980524.352004.351221. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Random Read - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P4K8K12K16K20K131081821617925

x265

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

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4K75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat71421283528.7324.7723.4231.3328.6431.7331.1622.9422.8624.3031.5731.321. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

JPEG XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: All75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat70140210280350320.16246.91278.06320.16332.73341.13317.83275.97284.19275.75318.90316.47

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Write - Clients: 25675F32AMD EPYC 75F3 32-Core110K220K330K440K550K507877367796

Rows: 10000 - Test: Random Write - Clients: 256

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

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010079.5388.3064.0172.0179.3576.6472.6380.0972.5472.3376.85

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: System V Message Passing75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core2M4M6M8M10M8711498.676498292.388962692.128704140.321. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 10075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat0.17330.34660.51990.69320.86650.710.560.690.740.690.750.770.730.730.700.730.731. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

x264

This is a multi-threaded test of the x264 video encoder run on the CPU with a choice of 1080p or 4K video input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2022-02-22Video Input: Bosphorus 1080p75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat60120180240300217.75193.65262.79205.05211.70200.65265.96221.25265.771. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -flto

JPEG XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: 175F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat112233445546.7036.9448.8850.7047.4050.4849.7848.5349.0248.5149.7149.83

x265

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

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080p75F328380 2p9554 TwoPAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010084.1874.4969.2884.0791.8870.5072.7693.5491.801. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: NUMA75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core2004006008001000925.07693.40904.20926.641. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat306090120150117.02120.3590.0995.97117.11106.5599.03111.4897.5096.23107.67

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 Compile75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat61218243026.3226.7020.2020.5026.1520.7220.7520.0920.0120.1320.4020.83

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Read - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P4080120160200189186142

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 275F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat102030405041.0244.4134.9433.7740.9533.4833.7534.6535.0734.5333.7033.841. (CXX) g++ options: -O3 -fPIC -lm

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Read - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P14002800420056007000507151896698

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 10000 - Test: Increment - Clients: 25675F32AMD EPYC 75F3 32-CoreEPYC 9654 2P20040060080010001139981864

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/. 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 2022.10.13Model: Bumper Beam75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010092.1086.3072.6469.9391.0170.4873.3976.4082.8375.7670.8277.93

FFmpeg

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Platform8380 2p9554 TwoP4080120160200175.80133.931. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Platform8380 2p9554 TwoP132639526543.0956.561. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 9075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat36912159.337.429.299.649.519.719.579.239.289.059.629.581. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Increment - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P20406080100907469

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha2_atonce8380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat4K8K12K16K20K20601.6115798.7620135.1015799.0920134.6920135.2020135.1419789.9620135.1720135.191. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: t1ha2_atonce8380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat61218243025.9026.8120.1025.7620.1020.1020.1020.1020.1020.101. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: t1ha2_atonce

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2040608010077.8579.6762.2662.0477.9361.1762.3063.5663.0863.1161.7362.271. (CXX) g++ options: -O3 -fPIC -lm

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 8075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat36912159.777.749.7210.0510.0210.089.959.579.689.499.959.941. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 8075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat36912159.457.529.539.779.599.799.689.299.299.329.709.681. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Increment - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P3K6K9K12K15K110711334114345

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 9075F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat36912159.687.749.649.989.8610.019.909.489.629.479.909.871. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

FFmpeg

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Video On Demand8380 2p9554 TwoP4080120160200174.66135.461. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Video On Demand8380 2p9554 TwoP132639526543.3755.921. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Upload8380 2p9554 TwoP306090120150116.4190.881. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Upload8380 2p9554 TwoP71421283521.6927.781. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: wyhash8380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat7K14K21K28K35K27859.5424475.9331245.4724509.4431198.5231218.0531248.9131247.3631197.9231201.201. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: wyhash8380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat51015202519.6318.1913.9818.0913.9713.9914.2213.9713.9713.971. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: wyhash

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: fasthash328380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat2K4K6K8K10K8453.996903.878803.736907.828803.528803.428803.408803.358802.198775.901. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: fasthash328380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat71421283528.8328.3522.0828.3322.0822.0822.0822.0822.0822.081. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: fasthash32

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: SHA3-2568380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat50100150200250208.31169.55216.08169.57216.08216.01216.15216.02216.13211.821. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: SHA3-2568380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat50010001500200025001876.362331.751829.032331.581829.751829.801829.211829.041830.341871.001. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: SHA3-256

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: MeowHash x86_64 AES-NI8380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat12K24K36K48K60K45558.1842531.5654175.0842535.2354220.2354190.5454144.8554192.2254202.2754212.961. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: MeowHash x86_64 AES-NI8380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat132639526551.4257.7445.1157.7345.1145.1145.1145.1145.1145.111. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: MeowHash x86_64 AES-NI

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: Spooky328380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat4K8K12K16K20K17883.7116145.6920577.0116149.3020581.5520577.1520576.7720580.2820582.7920582.571. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: Spooky328380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat91827364537.7034.2627.6234.2726.7026.8626.6627.6226.6826.661. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: Spooky32

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash1288380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat5K10K15K20K25K19206.2516608.9221167.1116609.2921167.6421167.1721167.2921167.4821168.0121166.211. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: FarmHash1288380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat142842567050.2261.3547.8261.3547.8347.8347.8347.8347.8347.831. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: FarmHash128

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash32 x86_64 AVX8380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat9K18K27K36K45K32568.1431226.8739794.1131226.5739796.1339795.9939795.4339794.9539795.0239795.041. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: FarmHash32 x86_64 AVX8380 2p9554 TwoP9654 PowerEPYC 9554EPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat81624324031.9133.6626.3933.6626.3926.3926.3926.3926.3926.391. (CXX) g++ options: -march=native -O3 -flto=auto -fno-fat-lto-objects

Hash: FarmHash32 x86_64 AVX

AMD EPYC 75F3 32-Core: The test quit with a non-zero exit status.

75F32: The test quit with a non-zero exit status.

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Read - Clients: 50075F32AMD EPYC 75F3 32-CoreEPYC 9654 2P130260390520650589529468

HammerDB - MariaDB

This is a MariaDB MySQL database server benchmark making use of the HammerDB benchmarking / load testing tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNew Orders Per Minute, More Is BetterHammerDB - MariaDB 10.9.3Virtual Users: 64 - Warehouses: 1008380 2p9554 TwoP10K20K30K40K50K46690371091. (CXX) g++ options: -fPIC -pie -fstack-protector -O2 -shared -lrt -lpthread -lz -ldl -lm -lstdc++

OpenBenchmarking.orgTransactions Per Minute, More Is BetterHammerDB - MariaDB 10.9.3Virtual Users: 64 - Warehouses: 1008380 2p9554 TwoP20K40K60K80K100K108494862311. (CXX) g++ options: -fPIC -pie -fstack-protector -O2 -shared -lrt -lpthread -lz -ldl -lm -lstdc++

spaCy

The spaCy library is an open-source solution for advanced neural language processing (NLP). The spaCy library leverages Python and is a leading neural language processing solution. This test profile times the spaCy CPU performance with various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgtokens/sec, More Is BetterspaCy 3.4.1Model: en_core_web_lg75F328380 2p9554 TwoP9654 PowerAMD EPYC 75F3 32-CoreEPYC 9654EPYC 9654 2PEPYC 9654 2P - Perf MEPYC 9654 2P RepeatEPYC 9654 PowerEPYC 9654 Repeat3K6K9K12K15K1364411140135401352613654135511348113415134671352713517

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Read - Clients: 50075F32AMD EPYC 75F3 32-CoreEPYC 9654 2P70K140K210K280K350K278941306869339482

FFmpeg

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Platform8380 2p9554 TwoP4080120160200187.34154.711. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Video On Demand8380 2p9554 TwoP112233445540.5649.111. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Platform8380 2p9554 TwoP112233445540.4448.961. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Video On Demand8380 2p9554 TwoP4080120160200186.74154.251. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Live8380 2p9554 TwoP71421283527.6822.891. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Live8380 2p9554 TwoP50100150200250182.45220.661. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Upload8380 2p9554 TwoP50100150200250239.53198.931. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx264 - Scenario: Upload8380 2p9554 TwoP369121510.5412.691. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Increment - Clients: 25675F32AMD EPYC 75F3 32-CoreEPYC 9654 2P30K60K90K120K150K129941141825155246

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 10000 - Test: Increment - Clients: 50075F32AMD EPYC 75F3 32-CoreEPYC 9654 2P400800120016002000170218211599

FFmpeg

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

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Live8380 2p9554 TwoP306090120150114.06127.921. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 5.1.2Encoder: libx265 - Scenario: Live8380 2p9554 TwoP102030405044.2739.481. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all queries performed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, Second Run75F328380 2pAMD EPYC 75F3 32-Core90180270360450401.99389.47402.41MIN: 43.7 / MAX: 20000MIN: 62.63 / MAX: 8571.43MIN: 43.07 / MAX: 200001. ClickHouse server version 22.5.4.19 (official build).

100M Rows Web Analytics Dataset, Second Run

9554 TwoP: The test run did not produce a result.

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Write - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P816243240343436

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all queries performed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, Third Run75F328380 2pAMD EPYC 75F3 32-Core90180270360450383.67382.84387.02MIN: 43.35 / MAX: 8571.43MIN: 64.66 / MAX: 8571.43MIN: 41.29 / MAX: 8571.431. ClickHouse server version 22.5.4.19 (official build).

100M Rows Web Analytics Dataset, Third Run

9554 TwoP: The test run did not produce a result.

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Write - Clients: 25675F32AMD EPYC 75F3 32-Core40080012001600200018711953

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Random Write - Clients: 175F32AMD EPYC 75F3 32-CoreEPYC 9654 2P6K12K18K24K30K270272695425907

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: IO_uring75F328380 2p9554 TwoPAMD EPYC 75F3 32-Core14K28K42K56K70K67158.1764705.5765946.5567389.931. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all queries performed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, First Run / Cold Cache75F328380 2pAMD EPYC 75F3 32-Core80160240320400377.33368.06376.51MIN: 37.85 / MAX: 10000MIN: 57.2 / MAX: 7500MIN: 37.24 / MAX: 100001. ClickHouse server version 22.5.4.19 (official build).

100M Rows Web Analytics Dataset, First Run / Cold Cache

9554 TwoP: The test run did not produce a result.

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Increment - Clients: 25675F32AMD EPYC 75F3 32-CoreEPYC 9654 2P400800120016002000202620621993

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Increment - Clients: 25675F32AMD EPYC 75F3 32-CoreEPYC 9654 2P30K60K90K120K150K125256123070127105

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Increment - Clients: 50075F32AMD EPYC 75F3 32-CoreEPYC 9654 2P30K60K90K120K150K153938154170149429

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Increment - Clients: 50075F32AMD EPYC 75F3 32-Core900180027003600450044094352

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Increment - Clients: 50075F32AMD EPYC 75F3 32-Core20K40K60K80K100K112527113983

Rows: 1000000 - Test: Increment - Clients: 500

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

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Scan - Clients: 25675F3213M26M39M52M65M58887584

Rows: 1000000 - Test: Scan - Clients: 256

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

AMD EPYC 75F3 32-Core: The test run did not produce a result.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Scan - Clients: 175F3230K60K90K120K150K131234

Rows: 1000000 - Test: Scan - Clients: 1

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

AMD EPYC 75F3 32-Core: The test run did not produce a result.

8380 2p: The test run did not produce a result.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 10000 - Test: Scan - Clients: 175F325K10K15K20K25K23753

Rows: 10000 - Test: Scan - Clients: 1

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

AMD EPYC 75F3 32-Core: The test run did not produce a result.

8380 2p: The test run did not produce a result.

Rows: 1000000 - Test: Scan - Clients: 500

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

AMD EPYC 75F3 32-Core: The test run did not produce a result.

75F32: The test run did not produce a result.

Rows: 10000 - Test: Scan - Clients: 500

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

AMD EPYC 75F3 32-Core: The test run did not produce a result.

75F32: The test run did not produce a result.

8380 2p: The test run did not produce a result.

Rows: 10000 - Test: Scan - Clients: 256

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

AMD EPYC 75F3 32-Core: The test run did not produce a result.

75F32: The test run did not produce a result.

8380 2p: The test run did not produce a result.

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all queries performed. Learn more via the OpenBenchmarking.org test page.

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

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

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

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

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

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

9654 Power: The test run did not produce a result.

9554 TwoP: The test run did not produce a result.

EPYC 9554: The test run did not produce a result.

188 Results Shown

Cpuminer-Opt
TensorFlow
Stress-NG
Neural Magic DeepSparse
TensorFlow:
  CPU - 32 - VGG-16
  CPU - 512 - GoogLeNet
Neural Magic DeepSparse
Stress-NG
Cpuminer-Opt
TensorFlow:
  CPU - 512 - ResNet-50
  CPU - 256 - GoogLeNet
  CPU - 16 - VGG-16
Neural Magic DeepSparse
Cpuminer-Opt
TensorFlow:
  CPU - 256 - ResNet-50
  CPU - 64 - AlexNet
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream
  NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream
OpenFOAM
Stress-NG
Cpuminer-Opt
Blender
Cpuminer-Opt
Blender:
  Pabellon Barcelona - CPU-Only
  Barbershop - CPU-Only
Cpuminer-Opt
Blender
7-Zip Compression
TensorFlow
Stress-NG:
  Glibc Qsort Data Sorting
  Glibc C String Functions
Blender
TensorFlow:
  CPU - 64 - ResNet-50
  CPU - 64 - GoogLeNet
Stress-NG
TensorFlow
Cpuminer-Opt
Stress-NG
Cpuminer-Opt
Stress-NG:
  CPU Stress
  Mutex
  Malloc
  Vector Math
  SENDFILE
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream
Stress-NG
Cpuminer-Opt
Timed Linux Kernel Compilation
TensorFlow
Cpuminer-Opt
HammerDB - MariaDB:
  64 - 500:
    Transactions Per Minute
    New Orders Per Minute
  64 - 250:
    New Orders Per Minute
Xmrig
HammerDB - MariaDB
Stress-NG:
  Context Switching
  MEMFD
TensorFlow
7-Zip Compression
Xmrig
Stress-NG
TensorFlow
Stress-NG
TensorFlow
OpenFOAM
OpenRadioss
Neural Magic DeepSparse:
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream:
    ms/batch
    items/sec
OpenRadioss
Stress-NG
Timed Node.js Compilation
Neural Magic DeepSparse:
  CV Detection,YOLOv5s COCO - Synchronous Single-Stream:
    ms/batch
    items/sec
OpenRadioss
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    ms/batch
    items/sec
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    ms/batch
    items/sec
Y-Cruncher:
  1B
  500M
x264
Timed Linux Kernel Compilation
spaCy
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream:
    ms/batch
    items/sec
Cpuminer-Opt
TensorFlow:
  CPU - 512 - VGG-16
  CPU - 256 - VGG-16
Cpuminer-Opt
TensorFlow:
  CPU - 512 - AlexNet
  CPU - 64 - VGG-16
OpenRadioss
Neural Magic DeepSparse:
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream
  NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream
  NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream
  NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream
Apache HBase
Timed Godot Game Engine Compilation
OpenFOAM
SMHasher
OpenFOAM
Stress-NG
Neural Magic DeepSparse
Apache HBase
libavif avifenc:
  6
  6, Lossless
Apache HBase:
  10000 - Increment - 1
  10000 - Rand Read - 256
Neural Magic DeepSparse
libavif avifenc
JPEG XL libjxl
Apache HBase
OpenFOAM
Apache HBase
x265
JPEG XL Decoding libjxl
Apache HBase
Neural Magic DeepSparse
Stress-NG
JPEG XL libjxl
x264
JPEG XL Decoding libjxl
x265
Stress-NG
Neural Magic DeepSparse
Timed Mesa Compilation
Apache HBase
libavif avifenc
Apache HBase:
  10000 - Rand Read - 1
  10000 - Increment - 256
OpenRadioss
FFmpeg:
  libx265 - Platform:
    Seconds
    FPS
JPEG XL libjxl
Apache HBase
SMHasher
libavif avifenc
JPEG XL libjxl:
  PNG - 80
  JPEG - 80
Apache HBase
JPEG XL libjxl
FFmpeg:
  libx265 - Video On Demand:
    Seconds
    FPS
  libx265 - Upload:
    Seconds
    FPS
SMHasher:
  wyhash
  fasthash32
  SHA3-256
  MeowHash x86_64 AES-NI
  Spooky32
  FarmHash128
  FarmHash32 x86_64 AVX
Apache HBase
HammerDB - MariaDB:
  64 - 100:
    New Orders Per Minute
    Transactions Per Minute
spaCy
Apache HBase
FFmpeg:
  libx264 - Platform
  libx264 - Video On Demand
  libx264 - Platform
  libx264 - Video On Demand
  libx264 - Live
  libx264 - Live
  libx264 - Upload
  libx264 - Upload
Apache HBase:
  10000 - Increment - 256
  10000 - Increment - 500
FFmpeg:
  libx265 - Live:
    FPS
    Seconds
ClickHouse
Apache HBase
ClickHouse
Apache HBase:
  10000 - Rand Write - 256
  10000 - Rand Write - 1
Stress-NG
ClickHouse
Apache HBase:
  1000000 - Increment - 256:
    Microseconds - Average Latency
    Rows Per Second
  10000 - Increment - 500:
    Rows Per Second
  1000000 - Increment - 500:
    Microseconds - Average Latency
    Rows Per Second
  1000000 - Scan - 256:
    Rows Per Second
  1000000 - Scan - 1:
    Rows Per Second
  10000 - Scan - 1:
    Rows Per Second

EPYC 9654 2P

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 15 October 2022 06:25 by user phoronix.

EPYC 9654 2P Repeat

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 15 October 2022 09:25 by user phoronix.

EPYC 9654 2P - Perf M

Processor: 2 x AMD EPYC 9654 96-Core @ 3.71GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 15 October 2022 11:52 by user phoronix.

EPYC 9654

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 15 October 2022 15:27 by user phoronix.

EPYC 9654 Repeat

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 15 October 2022 18:01 by user phoronix.

EPYC 9654 Power

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 16 October 2022 13:31 by user phoronix.

9654 Power

Processor: AMD EPYC 9654 96-Core @ 3.71GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 October 2022 07:48 by user phoronix.

AMD EPYC 75F3 32-Core

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 29 October 2022 13:43 by user phoronix.

75F32

Processor: AMD EPYC 75F3 32-Core @ 2.95GHz (32 Cores / 64 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 256GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710

OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 30 October 2022 04:10 by user phoronix.

9554 TwoP

Processor: 2 x AMD EPYC 9554 64-Core @ 3.76GHz (128 Cores / 256 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 1 November 2022 08:55 by user phoronix.

EPYC 9554

Processor: AMD EPYC 9554 64-Core @ 3.76GHz (64 Cores / 128 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110d
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 3 November 2022 08:42 by user phoronix.

8380 2p

Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Ice Lake IEH, Memory: 512GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP

OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0xd000375
Java Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)
Python Notes: Python 3.10.6
Security Notes: 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_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 5 November 2022 07:05 by user phoronix.