AMD EPYC 9684X 3D V-Cache Benchmark

AMD EPYC 9684X 96-Core testing by Michael Larabel for a future article. Various benchmarks conducted with the EPYC 9684X 1P and then repeated after disabling 3D V-Cache from the BIOS to see direct comparison of 3DV impact. Plus monitoring CPU thermal / power / frequency for future follow-up article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2307201-PTS-GENOAX3D86
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:

BLAS (Basic Linear Algebra Sub-Routine) Tests 3 Tests
C++ Boost Tests 3 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 6 Tests
C/C++ Compiler Tests 4 Tests
CPU Massive 13 Tests
Creator Workloads 9 Tests
Fortran Tests 7 Tests
Game Development 4 Tests
HPC - High Performance Computing 20 Tests
Linear Algebra 2 Tests
Machine Learning 5 Tests
Molecular Dynamics 7 Tests
MPI Benchmarks 7 Tests
Multi-Core 18 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 4 Tests
OpenMPI Tests 15 Tests
Programmer / Developer System Benchmarks 8 Tests
Python 2 Tests
Raytracing 2 Tests
Renderers 3 Tests
Scientific Computing 10 Tests
Software Defined Radio 2 Tests
Server CPU Tests 9 Tests
Texture Compression 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
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Multi-Way Comparison

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

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Default
July 17 2023
  1 Day, 46 Minutes
3DV Disabled
July 19 2023
  1 Day, 4 Hours, 53 Minutes
Invert Hiding All Results Option
  1 Day, 2 Hours, 50 Minutes
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):

Default

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101121
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

3DV Disabled

Processor: AMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 22.04, Kernel: 5.19.0-41-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 1024x768

AMD EPYC 9684X 3D V-Cache BenchmarkOpenBenchmarking.orgPhoronix Test SuiteAMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads)AMD Titanite_4G (RTI1007B BIOS)AMD Device 14a4768GB2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007ASPEEDBroadcom NetXtreme BCM5720 PCIeUbuntu 22.045.19.0-41-generic (x86_64)GNOME Shell 42.5X Server 1.21.1.41.3.224GCC 11.3.0ext41024x768ProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerVulkanCompilerFile-SystemScreen ResolutionAMD EPYC 9684X 3D V-Cache Benchmark PerformanceSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101121 - Python 3.10.6- 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

Default vs. 3DV Disabled ComparisonPhoronix Test SuiteBaseline+27.4%+27.4%+54.8%+54.8%+82.2%+82.2%tConvolve OpenMP - Degridding109.5%d.M.M.S - Execution Time104.6%Matrix 3D Math103.5%tConvolve MPI - Gridding76.2%6471.9%SP.C64.4%3263.6%AVL Tree62.3%V.D.F - CPU58.4%V.D.F - CPU58.3%tConvolve MPI - Degridding54.1%r2c - FFTW - float - 51233.7%d.L.M.S - Execution Time30.8%LU.C27.6%BT.C23.1%Eigen21.6%Monero - 1M19.8%c2c - FFTW - float - 51218.4%IS.D16.7%12816.2%CPU Cache15.5%MG.C14.7%CG.C14.1%c2c - FFTW - double - 25613.8%r2c - FFTW - double - 51213.6%tConvolve OpenMP - Gridding13.6%CPU - Numpy - 4194304 - Equation of State13.4%i.i.1.C.P.D13.3%BLAS13.2%r2c - FFTW - double - 25613.2%i.i.1.C.P.D12.8%Church Facade12%Malloc11.2%d.L.M.S - Mesh Time10.9%104 104 104 - 6010.9%conus 2.5km10.4%Matrix Math10.4%MPI CPU - water_GMX50_bare10.3%Pathtracer ISPC - Crown9.8%4009.5%Carbon Nanotube9.4%Pathtracer ISPC - Asian Dragon Obj9.1%2569%Pathtracer ISPC - Asian Dragon8.5%7.7%tConvolve MT - Degridding7.7%CPU - Numpy - 4194304 - Isoneutral Mixing7.5%Pipe7.3%Futex7.3%tConvolve MT - Gridding6.8%C75526.7%Exhaustive6.6%Fused Multiply-Add6.6%5006.6%Small6.2%allmodconfig6.2%N.Q.A.B.b.u.S.1.P - A.M.S6.1%Thorough6%N.Q.A.B.b.u.S.1.P - A.M.S6%S.F.P.R6%Time To Compile5.9%Ninja5.8%10005.7%2 - 4K - 1 - Path Tracer5.7%V.F.P5.6%2 - 4K - 16 - Path Tracer5.6%1 - 4K - 16 - Path Tracer5.5%3 - 4K - 1 - Path Tracer5.4%Lion5.3%3 - 4K - 32 - Path Tracer5.3%1 - 4K - 1 - Path Tracer5.2%2 - 4K - 32 - Path Tracer5%Barbershop - CPU-Only5%gravity_spheres_volume/dim_512/ao/real_time5%gravity_spheres_volume/dim_512/scivis/real_time5%3 - 4K - 16 - Path Tracer4.9%c2c - FFTW - double - 5124.9%1 - 4K - 32 - Path Tracer4.9%ATPase Simulation - 327,506 Atoms4.8%Time To Compile4.7%Memory Copying4.7%144 144 144 - 604.6%particle_volume/scivis/real_time4.3%EP.D4.3%particle_volume/ao/real_time4.3%Unix Makefiles4.2%Total Time4.1%P.D.F - CPU4%P.D.F - CPU3.9%C.S.9.P.Y.P - A.M.S3.8%Classroom - CPU-Only3.8%P.P.B.T.T3.7%C.S.9.P.Y.P - A.M.S3.7%Medium3.6%3.6%P.V.B.D.F - CPU3.6%gravity_spheres_volume/dim_512/pathtracer/real_time3.6%P.V.B.D.F - CPU3.5%3.5%Semaphores3.5%CPU - 512 - GoogLeNet3.3%Pabellon Barcelona - CPU-Only3.3%Fishy Cat - CPU-Only3.2%d.M.M.S - Mesh Time2.8%Streams2.7%160 160 160 - 602.6%CPU Stress2.6%Time To Compile2.6%P.D.F - CPU2.6%Mutex2.6%P.D.F - CPU2.4%192 - 256 - 5122.4%H.C.O2.4%Vector Math2.4%Wide Vector Math2.3%ASKAPOpenFOAMStress-NGASKAPlibxsmmNAS Parallel BenchmarkslibxsmmStress-NGOpenVINOOpenVINOASKAPHeFFTe - Highly Efficient FFT for ExascaleOpenFOAMNAS Parallel BenchmarksNAS Parallel BenchmarksLeelaChessZeroXmrigHeFFTe - Highly Efficient FFT for ExascaleNAS Parallel BenchmarkslibxsmmStress-NGNAS Parallel BenchmarksNAS Parallel BenchmarksHeFFTe - Highly Efficient FFT for ExascaleHeFFTe - Highly Efficient FFT for ExascaleASKAPPyHPC BenchmarksXcompact3d Incompact3dLeelaChessZeroHeFFTe - Highly Efficient FFT for ExascaleXcompact3d Incompact3dGoogle DracoStress-NGOpenFOAMHigh Performance Conjugate GradientWRFStress-NGGROMACSEmbreePalabosGPAWEmbreelibxsmmEmbreeLULESHASKAPPyHPC BenchmarksStress-NGStress-NGASKAPNgspiceASTC EncoderStress-NGPalabosminiFETimed Linux Kernel CompilationNeural Magic DeepSparseASTC EncoderNeural Magic DeepSparseACES DGEMMTimed Godot Game Engine CompilationTimed LLVM CompilationPalabosOSPRay StudioStress-NGOSPRay StudioOSPRay StudioOSPRay StudioGoogle DracoOSPRay StudioOSPRay StudioOSPRay StudioBlenderOSPRayOSPRayOSPRay StudioHeFFTe - Highly Efficient FFT for ExascaleOSPRay StudioNAMDTimed Node.js CompilationStress-NGHigh Performance Conjugate GradientOSPRayNAS Parallel BenchmarksOSPRayTimed LLVM CompilationStockfishOpenVINOOpenVINONeural Magic DeepSparseBlendersrsRAN ProjectNeural Magic DeepSparseASTC EncoderNumpy BenchmarkOpenVINOOSPRayOpenVINOAlgebraic Multi-Grid BenchmarkStress-NGTensorFlowBlenderBlenderOpenFOAMPETScHigh Performance Conjugate GradientStress-NGTimed PHP CompilationOpenVINOStress-NGOpenVINOLiquid-DSPASKAPStress-NGStress-NGDefault3DV Disabled

AMD EPYC 9684X 3D V-Cache Benchmarkopenfoam: drivaerFastback, Medium Mesh Size - Execution Timelibxsmm: 64npb: SP.Clibxsmm: 32stress-ng: AVL Treeopenvino: Vehicle Detection FP16 - CPUopenvino: Vehicle Detection FP16 - CPUaskap: tConvolve MPI - Degriddingheffte: r2c - FFTW - float - 512openfoam: drivaerFastback, Large Mesh Size - Execution Timenpb: LU.Cnpb: BT.Clczero: Eigenxmrig: Monero - 1Mheffte: c2c - FFTW - float - 512npb: IS.Dlibxsmm: 128stress-ng: CPU Cachenpb: MG.Cnpb: CG.Cheffte: c2c - FFTW - double - 256heffte: r2c - FFTW - double - 512pyhpc: CPU - Numpy - 4194304 - Equation of Statelczero: BLASheffte: r2c - FFTW - double - 256incompact3d: input.i3d 193 Cells Per Directiondraco: Church Facadestress-ng: Mallocopenfoam: drivaerFastback, Large Mesh Size - Mesh Timewrf: conus 2.5kmstress-ng: Matrix Mathgromacs: MPI CPU - water_GMX50_bareembree: Pathtracer ISPC - Crownpalabos: 400gpaw: Carbon Nanotubeembree: Pathtracer ISPC - Asian Dragon Objlibxsmm: 256embree: Pathtracer ISPC - Asian Dragonlulesh: askap: tConvolve MT - Degriddingpyhpc: CPU - Numpy - 4194304 - Isoneutral Mixingstress-ng: Pipestress-ng: Futexaskap: tConvolve MT - Griddingngspice: C7552astcenc: Exhaustivestress-ng: Fused Multiply-Addpalabos: 500minife: Smallbuild-linux-kernel: allmodconfigdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamastcenc: Thoroughdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streammt-dgemm: Sustained Floating-Point Ratebuild-godot: Time To Compilebuild-llvm: Ninjapalabos: 1000ospray-studio: 2 - 4K - 1 - Path Tracerstress-ng: Vector Floating Pointospray-studio: 2 - 4K - 16 - Path Tracerospray-studio: 1 - 4K - 16 - Path Tracerospray-studio: 3 - 4K - 1 - Path Tracerdraco: Lionospray-studio: 3 - 4K - 32 - Path Tracerospray-studio: 1 - 4K - 1 - Path Tracerospray-studio: 2 - 4K - 32 - Path Tracerblender: Barbershop - CPU-Onlyospray: gravity_spheres_volume/dim_512/ao/real_timeospray: gravity_spheres_volume/dim_512/scivis/real_timeospray-studio: 3 - 4K - 16 - Path Tracerheffte: c2c - FFTW - double - 512ospray-studio: 1 - 4K - 32 - Path Tracernamd: ATPase Simulation - 327,506 Atomsbuild-nodejs: Time To Compilestress-ng: Memory Copyinghpcg: 144 144 144 - 60ospray: particle_volume/scivis/real_timenpb: EP.Dospray: particle_volume/ao/real_timebuild-llvm: Unix Makefilesopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP32 - CPUdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamblender: Classroom - CPU-Onlysrsran: PUSCH Processor Benchmark, Throughput Totaldeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamastcenc: Mediumnumpy: openvino: Person Vehicle Bike Detection FP16 - CPUospray: gravity_spheres_volume/dim_512/pathtracer/real_timeopenvino: Person Vehicle Bike Detection FP16 - CPUamg: stress-ng: Semaphorestensorflow: CPU - 512 - GoogLeNetblender: Pabellon Barcelona - CPU-Onlyblender: Fishy Cat - CPU-Onlyopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timehpcg: 160 160 160 - 60stress-ng: CPU Stressbuild-php: Time To Compileopenvino: Person Detection FP16 - CPUstress-ng: Mutexopenvino: Person Detection FP16 - CPUliquid-dsp: 192 - 256 - 512askap: Hogbom Clean OpenMPstress-ng: Vector Mathstress-ng: Wide Vector Mathcloverleaf: Lagrangian-Eulerian Hydrodynamicsblender: BMW27 - CPU-Onlydeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamremhos: Sample Remap Exampledeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streambuild-gem5: Time To Compileopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUhpcg: 192 192 192 - 60ngspice: C2670petsc: Streamsstress-ng: Matrix 3D Mathaskap: tConvolve OpenMP - Degriddingaskap: tConvolve OpenMP - Griddingaskap: tConvolve MPI - Griddingstockfish: Total Timeincompact3d: input.i3d 129 Cells Per Directionhpcg: 104 104 104 - 60Default3DV Disabled181.017592455.4207614.701311.31665.5712.423860.6459791.1332.7428994.4651337910.64314777.901188469684.3153.8445696.512913.41397574.75137308.2759737.9487.0392134.8200.7679760190.1267.685286816043360348999.65585.1366211269.262418033.1311.793117.8261317.82034.774123.51353064.4143.995430715.32815603.21.57860302912.673985709.5613582.8100.8506.141176566577.12328.71354408.2202.557145.054456.7780329.977340.55318088.384112.887370.1891065257925.6117078169531261501140392105934341142.0326.792325.92722022367.8503340070.24733105.22732994.1824.614325.108810697.9025.1571184.04726.891766.54114.177740.5118408.1417.8584419.4265586.508.3626.51455732.752414283000223213939.86409.0249.6120.62108.3682323.8369212380.7633.44627.1149736118.061753.6312878666671212.17545725.733485374.3310.2916.28139.0157344.405310.293797.947160.0637137.6915672.068.4522.8332118.394272616.799416595.2855153.026625.673226.72972898922.1752161326.8250370.309171428.3126301.03801.61026.4019.672438.5638810.4248.85211763.369264731.92255679.10977058160.0129.9254881.222506.81209699.75119761.0752350.0476.4544118.6420.8708619167.9648.668864446768323948044.10649.0153812439.014378776.6510.690107.2730290.18938.054113.24232811.7132.771728515.35914487.71.69656210994.923716134.1412713.5107.5805.759871844167.06308.47251223.5215.052153.869953.5427311.201738.26342093.636119.453350.0911126244276.8018030178791329527842524111436073149.1925.519524.69672122264.7029356610.25909110.20231510.8023.524724.062210257.8124.1287191.72825.861835.23109.945342.0517748.8433.1377404.6610565.918.6625.59995538.582332743667215763708.61395.8551.2321.29111.4430523.2313206993.5934.31026.4348488923.411796.4212573666671183.56533171.173407854.7710.4816.57141.3071338.834810.457785.575561.0087139.7035603.348.5522.5732119.705265446.59288156.4626323.023436.741569.42855330402.4639547324.1985OpenBenchmarking.org

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 TimeDefault3DV Disabled80160240320400181.02370.311. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm
OpenBenchmarking.orgMegahertz, More Is BetterOpenFOAM 10CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3175.19 / Max: 3797Min: 2550 / Avg: 2992.79 / Max: 3771
OpenBenchmarking.orgWatts, Fewer Is BetterOpenFOAM 10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.95 / Avg: 301.19 / Max: 331.41Min: 32.97 / Avg: 311.68 / Max: 326.23
OpenBenchmarking.orgCelsius, Fewer Is BetterOpenFOAM 10CPU Temperature Monitor3DV DisabledDefault1428425670Min: 37.75 / Avg: 60.01 / Max: 71Min: 39.13 / Avg: 62.21 / Max: 71

libxsmm

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

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 64Default3DV Disabled5001000150020002500SE +/- 4.53, N = 7SE +/- 0.97, N = 62455.41428.31. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2
OpenBenchmarking.orgGFLOPS/s Per Watt, More Is Betterlibxsmm 2-1.17-3645M N K: 64Default3DV Disabled369121512.4026.708
OpenBenchmarking.orgMegahertz, More Is Betterlibxsmm 2-1.17-3645CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 2876.94 / Max: 3694Min: 2550 / Avg: 2736.27 / Max: 3693
OpenBenchmarking.orgWatts, Fewer Is Betterlibxsmm 2-1.17-3645CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 32.52 / Avg: 197.98 / Max: 337.23Min: 28.68 / Avg: 212.92 / Max: 312.051. Default: Approximate power consumption of 1782 Joules per run.2. 3DV Disabled: Approximate power consumption of 2555 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is Betterlibxsmm 2-1.17-3645CPU Temperature Monitor3DV DisabledDefault1224364860Min: 35.5 / Avg: 44.37 / Max: 62.25Min: 36.88 / Avg: 45.54 / Max: 51.25
OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 64Default3DV Disabled400800120016002000Min: 2434.2 / Avg: 2455.41 / Max: 2470.2Min: 1425.5 / Avg: 1428.32 / Max: 1432.41. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CDefault3DV Disabled40K80K120K160K200KSE +/- 1595.09, N = 6SE +/- 380.05, N = 4207614.70126301.031. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CDefault3DV Disabled20040060080010001009.13565.09
OpenBenchmarking.orgMegahertz, More Is BetterNAS Parallel Benchmarks 3.4CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3166.32 / Max: 3698Min: 2550 / Avg: 3037.34 / Max: 3753
OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 32.19 / Avg: 205.74 / Max: 320.66Min: 28.64 / Avg: 223.51 / Max: 311.551. Default: Approximate power consumption of 2400 Joules per run.2. 3DV Disabled: Approximate power consumption of 3632 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Temperature MonitorDefault3DV Disabled1122334455Min: 29.5 / Avg: 42.82 / Max: 53.63Min: 30.38 / Avg: 44.7 / Max: 54.38
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CDefault3DV Disabled40K80K120K160K200KMin: 202806.9 / Avg: 207614.7 / Max: 212234.42Min: 125628.3 / Avg: 126301.03 / Max: 127336.481. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

libxsmm

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

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 32Default3DV Disabled30060090012001500SE +/- 1.58, N = 8SE +/- 0.89, N = 61311.3801.61. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2
OpenBenchmarking.orgGFLOPS/s Per Watt, More Is Betterlibxsmm 2-1.17-3645M N K: 32Default3DV Disabled2468107.1334.129
OpenBenchmarking.orgMegahertz, More Is Betterlibxsmm 2-1.17-3645CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3105.22 / Max: 3694Min: 2550 / Avg: 2872.87 / Max: 3694
OpenBenchmarking.orgWatts, Fewer Is Betterlibxsmm 2-1.17-3645CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 33.12 / Avg: 183.83 / Max: 323.4Min: 29.16 / Avg: 194.16 / Max: 292.51. Default: Approximate power consumption of 1608 Joules per run.2. 3DV Disabled: Approximate power consumption of 2233 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is Betterlibxsmm 2-1.17-3645CPU Temperature Monitor3DV DisabledDefault1122334455Min: 38.75 / Avg: 45.51 / Max: 53.13Min: 39.88 / Avg: 46.95 / Max: 52.5
OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 32Default3DV Disabled2004006008001000Min: 1302.2 / Avg: 1311.31 / Max: 1316.3Min: 798.5 / Avg: 801.6 / Max: 804.51. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: AVL TreeDefault3DV Disabled400800120016002000SE +/- 0.40, N = 3SE +/- 0.16, N = 31665.571026.401. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: AVL TreeDefault3DV Disabled1.24312.48623.72934.97246.21555.5253.712
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 3024 / Avg: 3227.18 / Max: 3708Min: 3026 / Avg: 3127.82 / Max: 3700
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 107.49 / Avg: 276.5 / Max: 302.89Min: 113.33 / Avg: 301.48 / Max: 333.551. 3DV Disabled: Approximate power consumption of 9401 Joules per run.2. Default: Approximate power consumption of 10250 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1224364860Min: 40.75 / Avg: 54.64 / Max: 59.63Min: 40.75 / Avg: 57.62 / Max: 63
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: AVL TreeDefault3DV Disabled30060090012001500Min: 1665.14 / Avg: 1665.57 / Max: 1666.38Min: 1026.13 / Avg: 1026.4 / Max: 1026.671. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUDefault3DV Disabled510152025SE +/- 0.00, N = 3SE +/- 0.14, N = 1412.4219.67MIN: 5.8 / MAX: 47.23MIN: 5.14 / MAX: 51.621. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgMegahertz, More Is BetterOpenVINO 2022.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2396 / Avg: 2797.51 / Max: 3696Min: 2459 / Avg: 2505.37 / Max: 3699
OpenBenchmarking.orgWatts, Fewer Is BetterOpenVINO 2022.3CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.43 / Avg: 297.51 / Max: 326.48Min: 33.69 / Avg: 310.21 / Max: 333.371. 3DV Disabled: Approximate power consumption of 19062 Joules per run.2. Default: Approximate power consumption of 19957 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOpenVINO 2022.3CPU Temperature Monitor3DV DisabledDefault1224364860Min: 37.75 / Avg: 53.69 / Max: 63.13Min: 43.63 / Avg: 60.31 / Max: 63.5
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUDefault3DV Disabled510152025Min: 12.42 / Avg: 12.42 / Max: 12.42Min: 17.92 / Avg: 19.67 / Max: 20.011. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUDefault3DV Disabled8001600240032004000SE +/- 0.30, N = 3SE +/- 18.66, N = 143860.642438.561. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUDefault3DV Disabled7001400210028003500Min: 3860.14 / Avg: 3860.64 / Max: 3861.17Min: 2395.39 / Avg: 2438.56 / Max: 2675.151. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingDefault3DV Disabled13K26K39K52K65KSE +/- 380.83, N = 3SE +/- 363.62, N = 659791.138810.41. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingDefault3DV Disabled10K20K30K40K50KMin: 59410.3 / Avg: 59791.13 / Max: 60552.8Min: 37936.7 / Avg: 38810.37 / Max: 40368.61. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

HeFFTe - Highly Efficient FFT for Exascale

HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: float - X Y Z: 512Default3DV Disabled70140210280350SE +/- 1.86, N = 6SE +/- 0.81, N = 5332.74248.851. (CXX) g++ options: -O3
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: float - X Y Z: 512Default3DV Disabled0.40230.80461.20691.60922.01151.7881.291
OpenBenchmarking.orgMegahertz, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3221.52 / Max: 3816Min: 2550 / Avg: 3025.14 / Max: 3700
OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 17.79 / Avg: 186.14 / Max: 315.64Min: 28.2 / Avg: 192.71 / Max: 299.531. Default: Approximate power consumption of 2017 Joules per run.2. 3DV Disabled: Approximate power consumption of 2428 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Temperature Monitor3DV DisabledDefault1122334455Min: 27.13 / Avg: 40.4 / Max: 53.88Min: 28.38 / Avg: 41.2 / Max: 52.88
OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: float - X Y Z: 512Default3DV Disabled60120180240300Min: 328.03 / Avg: 332.74 / Max: 340.29Min: 246.25 / Avg: 248.85 / Max: 250.561. (CXX) g++ options: -O3

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, Large Mesh Size - Execution TimeDefault3DV Disabled3K6K9K12K15K8994.4711763.371. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm
OpenBenchmarking.orgMegahertz, More Is BetterOpenFOAM 10CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3366.28 / Max: 3755Min: 2550 / Avg: 3300.55 / Max: 3866
OpenBenchmarking.orgWatts, Fewer Is BetterOpenFOAM 10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.37 / Avg: 290.38 / Max: 332.84Min: 32.82 / Avg: 298.03 / Max: 329.05
OpenBenchmarking.orgCelsius, Fewer Is BetterOpenFOAM 10CPU Temperature Monitor3DV DisabledDefault1530456075Min: 37.25 / Avg: 57.13 / Max: 78.13Min: 39.13 / Avg: 58.29 / Max: 77.63

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CDefault3DV Disabled70K140K210K280K350KSE +/- 1381.97, N = 6SE +/- 1268.32, N = 5337910.64264731.921. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CDefault3DV Disabled4008001200160020001850.961380.70
OpenBenchmarking.orgMegahertz, More Is BetterNAS Parallel Benchmarks 3.4CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3267.13 / Max: 3703Min: 2550 / Avg: 3247.2 / Max: 3701
OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption MonitorDefault3DV Disabled50100150200250Min: 32.28 / Avg: 182.56 / Max: 286.73Min: 28.1 / Avg: 191.74 / Max: 281.671. Default: Approximate power consumption of 1947 Joules per run.2. 3DV Disabled: Approximate power consumption of 2378 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Temperature MonitorDefault3DV Disabled1224364860Min: 32.75 / Avg: 45.35 / Max: 56.88Min: 33.13 / Avg: 47.37 / Max: 61.13
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CDefault3DV Disabled60K120K180K240K300KMin: 334059.16 / Avg: 337910.64 / Max: 343221.71Min: 261257.11 / Avg: 264731.92 / Max: 268907.521. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CDefault3DV Disabled70K140K210K280K350KSE +/- 1672.02, N = 5SE +/- 1397.51, N = 4314777.90255679.101. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CDefault3DV Disabled300600900120015001450.511150.46
OpenBenchmarking.orgMegahertz, More Is BetterNAS Parallel Benchmarks 3.4CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3115.82 / Max: 3705Min: 2550 / Avg: 3048.86 / Max: 3764
OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 32.98 / Avg: 217.01 / Max: 307.43Min: 29.2 / Avg: 222.24 / Max: 307.331. Default: Approximate power consumption of 3038 Joules per run.2. 3DV Disabled: Approximate power consumption of 3556 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Temperature MonitorDefault3DV Disabled1122334455Min: 35.5 / Avg: 49.34 / Max: 56.88Min: 36 / Avg: 49.43 / Max: 58
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CDefault3DV Disabled50K100K150K200K250KMin: 308849.6 / Avg: 314777.9 / Max: 318436.3Min: 252236.14 / Avg: 255679.1 / Max: 259057.231. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

LeelaChessZero

LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: EigenDefault3DV Disabled3K6K9K12K15KSE +/- 72.53, N = 3SE +/- 103.20, N = 51188497701. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.28Backend: EigenDefault3DV Disabled102030405045.3338.64
OpenBenchmarking.orgMegahertz, More Is BetterLeelaChessZero 0.28CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3454.62 / Max: 3863Min: 2550 / Avg: 3427.01 / Max: 3725
OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.28CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 29.35 / Avg: 252.82 / Max: 297.93Min: 33.8 / Avg: 262.16 / Max: 305.851. 3DV Disabled: Approximate power consumption of 93645 Joules per run.2. Default: Approximate power consumption of 96651 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterLeelaChessZero 0.28CPU Temperature Monitor3DV DisabledDefault1428425670Min: 37.75 / Avg: 57.71 / Max: 71.13Min: 38.25 / Avg: 59.54 / Max: 70.63
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: EigenDefault3DV Disabled2K4K6K8K10KMin: 11784 / Avg: 11884 / Max: 12025Min: 9507 / Avg: 9770 / Max: 100691. (CXX) g++ options: -flto -pthread

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: 1MDefault3DV Disabled15K30K45K60K75KSE +/- 83.67, N = 4SE +/- 265.77, N = 369684.358160.01. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s Per Watt, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1MDefault3DV Disabled50100150200250246.10208.70
OpenBenchmarking.orgMegahertz, More Is BetterXmrig 6.18.1CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 2821.57 / Max: 3696Min: 2550 / Avg: 2814.64 / Max: 3699
OpenBenchmarking.orgWatts, Fewer Is BetterXmrig 6.18.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.32 / Avg: 278.68 / Max: 341.51Min: 32.46 / Avg: 283.15 / Max: 362.41. 3DV Disabled: Approximate power consumption of 6038 Joules per run.2. Default: Approximate power consumption of 5380 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterXmrig 6.18.1CPU Temperature Monitor3DV DisabledDefault1224364860Min: 35 / Avg: 51.35 / Max: 58Min: 36 / Avg: 53.72 / Max: 61.25
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1MDefault3DV Disabled12K24K36K48K60KMin: 69454.1 / Avg: 69684.28 / Max: 69832.4Min: 57630.2 / Avg: 58159.97 / Max: 58462.41. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

HeFFTe - Highly Efficient FFT for Exascale

HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: float - X Y Z: 512Default3DV Disabled306090120150SE +/- 0.85, N = 4SE +/- 0.17, N = 4153.84129.931. (CXX) g++ options: -O3
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: float - X Y Z: 512Default3DV Disabled0.15590.31180.46770.62360.77950.6930.602
OpenBenchmarking.orgMegahertz, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3334.58 / Max: 3751Min: 2550 / Avg: 3209.51 / Max: 3701
OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.22 / Avg: 215.71 / Max: 297.53Min: 31.75 / Avg: 221.88 / Max: 309.051. 3DV Disabled: Approximate power consumption of 4152 Joules per run.2. Default: Approximate power consumption of 3827 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Temperature Monitor3DV DisabledDefault1122334455Min: 28.88 / Avg: 43.98 / Max: 57.38Min: 29.88 / Avg: 45.23 / Max: 57.63
OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: float - X Y Z: 512Default3DV Disabled306090120150Min: 151.29 / Avg: 153.84 / Max: 154.77Min: 129.47 / Avg: 129.92 / Max: 130.21. (CXX) g++ options: -O3

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DDefault3DV Disabled12002400360048006000SE +/- 29.57, N = 5SE +/- 36.66, N = 55696.514881.221. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DDefault3DV Disabled81624324033.2428.24
OpenBenchmarking.orgMegahertz, More Is BetterNAS Parallel Benchmarks 3.4CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3334.64 / Max: 3720Min: 2550 / Avg: 3323.65 / Max: 3757
OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption MonitorDefault3DV Disabled50100150200250Min: 32.22 / Avg: 171.38 / Max: 288.84Min: 28.51 / Avg: 172.83 / Max: 288.251. Default: Approximate power consumption of 2159 Joules per run.2. 3DV Disabled: Approximate power consumption of 2316 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Temperature MonitorDefault3DV Disabled1224364860Min: 32.75 / Avg: 43.64 / Max: 58Min: 33.13 / Avg: 44.95 / Max: 60.63
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DDefault3DV Disabled10002000300040005000Min: 5617.45 / Avg: 5696.51 / Max: 5762.12Min: 4780.43 / Avg: 4881.22 / Max: 4959.841. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

libxsmm

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

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 128Default3DV Disabled6001200180024003000SE +/- 27.97, N = 9SE +/- 6.05, N = 32913.42506.81. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2
OpenBenchmarking.orgGFLOPS/s Per Watt, More Is Betterlibxsmm 2-1.17-3645M N K: 128Default3DV Disabled369121510.2808.638
OpenBenchmarking.orgMegahertz, More Is Betterlibxsmm 2-1.17-3645CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2540 / Avg: 3564.54 / Max: 3682Min: 2550 / Avg: 3490.23 / Max: 3698
OpenBenchmarking.orgWatts, Fewer Is Betterlibxsmm 2-1.17-3645CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 33.79 / Avg: 283.4 / Max: 346.01Min: 29.24 / Avg: 290.21 / Max: 328.251. Default: Approximate power consumption of 134049 Joules per run.2. 3DV Disabled: Approximate power consumption of 131753 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is Betterlibxsmm 2-1.17-3645CPU Temperature MonitorDefault3DV Disabled1326395265Min: 39.38 / Avg: 63.28 / Max: 66.25Min: 36 / Avg: 64.1 / Max: 66.75
OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 128Default3DV Disabled5001000150020002500Min: 2840.3 / Avg: 2913.37 / Max: 3065Min: 2495.7 / Avg: 2506.8 / Max: 2516.51. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: CPU CacheDefault3DV Disabled300K600K900K1200K1500KSE +/- 11233.16, N = 3SE +/- 17248.98, N = 31397574.751209699.751. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: CPU CacheDefault3DV Disabled2K4K6K8K10K11493.2410174.76
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2995 / Avg: 3680.1 / Max: 3736Min: 3298 / Avg: 3661.46 / Max: 3740
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 104.37 / Avg: 118.89 / Max: 189.4Min: 108.97 / Avg: 121.6 / Max: 276.391. 3DV Disabled: Approximate power consumption of 4201 Joules per run.2. Default: Approximate power consumption of 4378 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1122334455Min: 39.38 / Avg: 42.58 / Max: 50Min: 38.5 / Avg: 42.91 / Max: 54.5
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: CPU CacheDefault3DV Disabled200K400K600K800K1000KMin: 1378221.61 / Avg: 1397574.75 / Max: 1417132.81Min: 1183235.11 / Avg: 1209699.75 / Max: 1242097.131. (CXX) g++ options: -O2 -std=gnu99 -lc

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CDefault3DV Disabled30K60K90K120K150KSE +/- 1630.26, N = 15SE +/- 1316.62, N = 15137308.27119761.071. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CDefault3DV Disabled300600900120015001400.061191.49
OpenBenchmarking.orgMegahertz, More Is BetterNAS Parallel Benchmarks 3.4CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3075.48 / Max: 3716Min: 2550 / Avg: 2980.45 / Max: 3706
OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption MonitorDefault3DV Disabled50100150200250Min: 16.08 / Avg: 97.44 / Max: 284.81Min: 27.96 / Avg: 101.32 / Max: 285.031. Default: Approximate power consumption of 617 Joules per run.2. 3DV Disabled: Approximate power consumption of 621 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Temperature MonitorDefault3DV Disabled1122334455Min: 30.88 / Avg: 36.01 / Max: 52.5Min: 31.25 / Avg: 37.98 / Max: 56.5
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CDefault3DV Disabled20K40K60K80K100KMin: 124495.38 / Avg: 137308.27 / Max: 148883.34Min: 109428.49 / Avg: 119761.07 / Max: 132276.211. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CDefault3DV Disabled13K26K39K52K65KSE +/- 470.01, N = 10SE +/- 512.83, N = 1559737.9452350.041. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CDefault3DV Disabled100200300400500453.90366.48
OpenBenchmarking.orgMegahertz, More Is BetterNAS Parallel Benchmarks 3.4CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 3037.77 / Max: 3739Min: 2550 / Avg: 3036.6 / Max: 3757
OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption MonitorDefault3DV Disabled50100150200250Min: 32.67 / Avg: 131.61 / Max: 291.44Min: 28.45 / Avg: 142.85 / Max: 295.931. Default: Approximate power consumption of 921 Joules per run.2. 3DV Disabled: Approximate power consumption of 1076 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Temperature MonitorDefault3DV Disabled1224364860Min: 34.5 / Avg: 42.45 / Max: 57.5Min: 35.5 / Avg: 43.82 / Max: 59
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CDefault3DV Disabled10K20K30K40K50KMin: 57373.28 / Avg: 59737.94 / Max: 62290.53Min: 49585.46 / Avg: 52350.04 / Max: 57272.311. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

HeFFTe - Highly Efficient FFT for Exascale

HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 256Default3DV Disabled20406080100SE +/- 0.52, N = 9SE +/- 0.50, N = 987.0476.451. (CXX) g++ options: -O3
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 256Default3DV Disabled0.14650.2930.43950.5860.73250.6510.556
OpenBenchmarking.orgMegahertz, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 3010.57 / Max: 3810Min: 2550 / Avg: 2951.04 / Max: 3707
OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 31.58 / Avg: 133.77 / Max: 307.49Min: 27.63 / Avg: 137.54 / Max: 309.521. Default: Approximate power consumption of 951 Joules per run.2. 3DV Disabled: Approximate power consumption of 1039 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Temperature Monitor3DV DisabledDefault1020304050Min: 27.13 / Avg: 36.18 / Max: 49.75Min: 28.38 / Avg: 37.29 / Max: 50.75
OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 256Default3DV Disabled20406080100Min: 85.11 / Avg: 87.04 / Max: 89.4Min: 74.2 / Avg: 76.45 / Max: 78.491. (CXX) g++ options: -O3

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 512Default3DV Disabled306090120150SE +/- 0.58, N = 4SE +/- 0.47, N = 3134.82118.641. (CXX) g++ options: -O3
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 512Default3DV Disabled0.13070.26140.39210.52280.65350.5810.536
OpenBenchmarking.orgMegahertz, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3351.52 / Max: 3695Min: 2550 / Avg: 3243.96 / Max: 3839
OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.63 / Avg: 221.19 / Max: 299.05Min: 31.74 / Avg: 232.12 / Max: 306.591. 3DV Disabled: Approximate power consumption of 4719 Joules per run.2. Default: Approximate power consumption of 4410 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Temperature Monitor3DV DisabledDefault1224364860Min: 28.38 / Avg: 44.14 / Max: 57Min: 29.5 / Avg: 46.1 / Max: 58.25
OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 512Default3DV Disabled306090120150Min: 133.74 / Avg: 134.82 / Max: 135.83Min: 117.72 / Avg: 118.64 / Max: 119.261. (CXX) g++ options: -O3

PyHPC Benchmarks

PyHPC-Benchmarks is a suite of Python high performance computing benchmarks for execution on CPUs and GPUs using various popular Python HPC libraries. The PyHPC CPU-based benchmarks focus on sequential CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPyHPC Benchmarks 3.0Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of StateDefault3DV Disabled0.19580.39160.58740.78320.979SE +/- 0.004, N = 3SE +/- 0.001, N = 30.7670.870
OpenBenchmarking.orgMegahertz, More Is BetterPyHPC Benchmarks 3.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3591.44 / Max: 3700Min: 2550 / Avg: 3590.06 / Max: 3729
OpenBenchmarking.orgWatts, Fewer Is BetterPyHPC Benchmarks 3.0CPU Power Consumption Monitor3DV DisabledDefault20406080100Min: 27.82 / Avg: 100.44 / Max: 128.11Min: 32.18 / Avg: 103.69 / Max: 131.821. 3DV Disabled: Approximate power consumption of 3482 Joules per run.2. Default: Approximate power consumption of 3249 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterPyHPC Benchmarks 3.0CPU Temperature Monitor3DV DisabledDefault816243240Min: 27.13 / Avg: 33.07 / Max: 34.75Min: 29.5 / Avg: 35.27 / Max: 38.5
OpenBenchmarking.orgSeconds, Fewer Is BetterPyHPC Benchmarks 3.0Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of StateDefault3DV Disabled246810Min: 0.76 / Avg: 0.77 / Max: 0.77Min: 0.87 / Avg: 0.87 / Max: 0.87

LeelaChessZero

LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: BLASDefault3DV Disabled2K4K6K8K10KSE +/- 103.04, N = 5SE +/- 93.12, N = 4976086191. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.28Backend: BLASDefault3DV Disabled91827364538.5134.91
OpenBenchmarking.orgMegahertz, More Is BetterLeelaChessZero 0.28CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3466.94 / Max: 3847Min: 2550 / Avg: 3451.79 / Max: 3759
OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.28CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 28.84 / Avg: 246.89 / Max: 292.66Min: 32.13 / Avg: 253.45 / Max: 299.71. 3DV Disabled: Approximate power consumption of 91165 Joules per run.2. Default: Approximate power consumption of 93373 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterLeelaChessZero 0.28CPU Temperature Monitor3DV DisabledDefault1428425670Min: 33.63 / Avg: 57.23 / Max: 68.63Min: 33.13 / Avg: 58.62 / Max: 73.38
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: BLASDefault3DV Disabled2K4K6K8K10KMin: 9467 / Avg: 9759.8 / Max: 10107Min: 8447 / Avg: 8619 / Max: 88751. (CXX) g++ options: -flto -pthread

HeFFTe - Highly Efficient FFT for Exascale

HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 256Default3DV Disabled4080120160200SE +/- 1.48, N = 15SE +/- 1.47, N = 15190.13167.961. (CXX) g++ options: -O3
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 256Default3DV Disabled0.42570.85141.27711.70282.12851.8921.662
OpenBenchmarking.orgMegahertz, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3100.91 / Max: 3710Min: 2550 / Avg: 2847.35 / Max: 3806
OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 21.46 / Avg: 100.46 / Max: 308.5Min: 27.96 / Avg: 101.03 / Max: 283.651. Default: Approximate power consumption of 589 Joules per run.2. 3DV Disabled: Approximate power consumption of 606 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Temperature Monitor3DV DisabledDefault1020304050Min: 28.13 / Avg: 33.54 / Max: 45.5Min: 29.5 / Avg: 34.41 / Max: 47
OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 256Default3DV Disabled306090120150Min: 181.92 / Avg: 190.13 / Max: 200.56Min: 161.01 / Avg: 167.96 / Max: 177.711. (CXX) g++ options: -O3

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionDefault3DV Disabled246810SE +/- 0.06675033, N = 5SE +/- 0.03638771, N = 57.685286818.668864441. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
OpenBenchmarking.orgMegahertz, More Is BetterXcompact3d Incompact3d 2021-03-11CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3106.6 / Max: 3738Min: 2550 / Avg: 3054.46 / Max: 3698
OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 16.21 / Avg: 205.88 / Max: 310.52Min: 28.55 / Avg: 207.33 / Max: 302.651. Default: Approximate power consumption of 2594 Joules per run.2. 3DV Disabled: Approximate power consumption of 2778 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Temperature Monitor3DV DisabledDefault1122334455Min: 30.88 / Avg: 44.26 / Max: 54.38Min: 32.75 / Avg: 44.42 / Max: 53.5
OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionDefault3DV Disabled3691215Min: 7.59 / Avg: 7.69 / Max: 7.94Min: 8.55 / Avg: 8.67 / Max: 8.761. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Church FacadeDefault3DV Disabled15003000450060007500SE +/- 3.90, N = 6SE +/- 17.42, N = 6604367681. (CXX) g++ options: -O3
OpenBenchmarking.orgMegahertz, More Is BetterGoogle Draco 1.5.6CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3446.59 / Max: 3697Min: 2550 / Avg: 3391.49 / Max: 3698
OpenBenchmarking.orgWatts, Fewer Is BetterGoogle Draco 1.5.6CPU Power Consumption Monitor3DV DisabledDefault20406080100Min: 28.02 / Avg: 79.4 / Max: 103.63Min: 31.38 / Avg: 80.59 / Max: 107.391. 3DV Disabled: Approximate power consumption of 966 Joules per run.2. Default: Approximate power consumption of 900 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterGoogle Draco 1.5.6CPU Temperature MonitorDefault3DV Disabled816243240Min: 29.5 / Avg: 32.87 / Max: 34.75Min: 29.5 / Avg: 33.51 / Max: 35.25
OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Church FacadeDefault3DV Disabled12002400360048006000Min: 6030 / Avg: 6043.33 / Max: 6053Min: 6711 / Avg: 6768.33 / Max: 68181. (CXX) g++ options: -O3

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: MallocDefault3DV Disabled80M160M240M320M400MSE +/- 494466.73, N = 3SE +/- 35157.83, N = 3360348999.65323948044.101. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: MallocDefault3DV Disabled300K600K900K1200K1500K1204078.981126317.75
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3029.77 / Max: 3696Min: 2550 / Avg: 2907.67 / Max: 3694
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.22 / Avg: 287.62 / Max: 332.44Min: 33.03 / Avg: 299.27 / Max: 345.141. 3DV Disabled: Approximate power consumption of 9971 Joules per run.2. Default: Approximate power consumption of 10375 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1326395265Min: 37.75 / Avg: 59 / Max: 65Min: 38.25 / Avg: 60.02 / Max: 66
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: MallocDefault3DV Disabled60M120M180M240M300MMin: 359379546.43 / Avg: 360348999.65 / Max: 361002878.18Min: 323912551.07 / Avg: 323948044.1 / Max: 324018358.71. (CXX) g++ options: -O2 -std=gnu99 -lc

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, Large Mesh Size - Mesh TimeDefault3DV Disabled140280420560700585.14649.021. (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

WRF

WRF, the Weather Research and Forecasting Model, is a "next-generation mesoscale numerical weather prediction system designed for both atmospheric research and operational forecasting applications. It features two dynamical cores, a data assimilation system, and a software architecture supporting parallel computation and system extensibility." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWRF 4.2.2Input: conus 2.5kmDefault3DV Disabled3K6K9K12K15K11269.2612439.011. (F9X) gfortran options: -O2 -ftree-vectorize -funroll-loops -ffree-form -fconvert=big-endian -frecord-marker=4 -fallow-invalid-boz -lesmf_time -lwrfio_nf -lnetcdff -lnetcdf -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
OpenBenchmarking.orgMegahertz, More Is BetterWRF 4.2.2CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3330.48 / Max: 3752Min: 2550 / Avg: 3275.22 / Max: 3754
OpenBenchmarking.orgWatts, Fewer Is BetterWRF 4.2.2CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 31.21 / Avg: 285.69 / Max: 327.79Min: 35.91 / Avg: 294.72 / Max: 325.55
OpenBenchmarking.orgCelsius, Fewer Is BetterWRF 4.2.2CPU Temperature Monitor3DV DisabledDefault1428425670Min: 34.5 / Avg: 56.55 / Max: 71.5Min: 33.75 / Avg: 59.58 / Max: 74.75

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Matrix MathDefault3DV Disabled90K180K270K360K450KSE +/- 67.17, N = 3SE +/- 16.98, N = 3418033.13378776.651. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Matrix Math3DV DisabledDefault300600900120015001478.721428.83
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2705.19 / Max: 2731Min: 2465 / Avg: 2513.75 / Max: 3696
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 31.1 / Avg: 256.15 / Max: 289.76Min: 34.72 / Avg: 292.57 / Max: 332.411. 3DV Disabled: Approximate power consumption of 8795 Joules per run.2. Default: Approximate power consumption of 9947 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1224364860Min: 39.13 / Avg: 54.04 / Max: 57.38Min: 39.63 / Avg: 57.74 / Max: 61.75
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Matrix MathDefault3DV Disabled70K140K210K280K350KMin: 417898.8 / Avg: 418033.13 / Max: 418102.42Min: 378742.78 / Avg: 378776.65 / Max: 378795.641. (CXX) g++ options: -O2 -std=gnu99 -lc

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2023Implementation: MPI CPU - Input: water_GMX50_bareDefault3DV Disabled3691215SE +/- 0.01, N = 3SE +/- 0.01, N = 311.7910.691. (CXX) g++ options: -O3
OpenBenchmarking.orgNs Per Day Per Watt, More Is BetterGROMACS 2023Implementation: MPI CPU - Input: water_GMX50_bareDefault3DV Disabled0.01010.02020.03030.04040.05050.0450.042
OpenBenchmarking.orgMegahertz, More Is BetterGROMACS 2023CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3036.39 / Max: 3733Min: 2550 / Avg: 2992.28 / Max: 3700
OpenBenchmarking.orgWatts, Fewer Is BetterGROMACS 2023CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.5 / Avg: 252.54 / Max: 322.8Min: 32.25 / Avg: 260.11 / Max: 318.871. 3DV Disabled: Approximate power consumption of 7829 Joules per run.2. Default: Approximate power consumption of 7630 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterGROMACS 2023CPU Temperature Monitor3DV DisabledDefault1326395265Min: 32.25 / Avg: 50.87 / Max: 66.75Min: 32.63 / Avg: 51.67 / Max: 68.25
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2023Implementation: MPI CPU - Input: water_GMX50_bareDefault3DV Disabled3691215Min: 11.77 / Avg: 11.79 / Max: 11.81Min: 10.67 / Avg: 10.69 / Max: 10.711. (CXX) g++ options: -O3

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: CrownDefault3DV Disabled306090120150SE +/- 0.08, N = 7SE +/- 0.10, N = 6117.83107.27MIN: 114.92 / MAX: 122.62MIN: 104.48 / MAX: 111.58
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: CrownDefault3DV Disabled0.13120.26240.39360.52480.6560.5830.536
OpenBenchmarking.orgMegahertz, More Is BetterEmbree 4.1CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2775.38 / Max: 3697Min: 2550 / Avg: 2760.86 / Max: 3696
OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 4.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.77 / Avg: 200.22 / Max: 319Min: 32.88 / Avg: 201.95 / Max: 338.571. 3DV Disabled: Approximate power consumption of 2102 Joules per run.2. Default: Approximate power consumption of 1962 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterEmbree 4.1CPU Temperature Monitor3DV DisabledDefault1122334455Min: 37.75 / Avg: 46.85 / Max: 53.88Min: 38.25 / Avg: 47.94 / Max: 55.75
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: CrownDefault3DV Disabled20406080100Min: 117.44 / Avg: 117.83 / Max: 118.09Min: 106.92 / Avg: 107.27 / Max: 107.57

Palabos

The Palabos library is a framework for general purpose Computational Fluid Dynamics (CFD). Palabos uses a kernel based on the Lattice Boltzmann method. This test profile uses the Palabos MPI-based Cavity3D benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMega Site Updates Per Second, More Is BetterPalabos 2.3Grid Size: 400Default3DV Disabled70140210280350SE +/- 2.86, N = 12SE +/- 0.38, N = 3317.82290.191. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm
OpenBenchmarking.orgMega Site Updates Per Second Per Watt, More Is BetterPalabos 2.3Grid Size: 400Default3DV Disabled0.26840.53680.80521.07361.3421.1931.135
OpenBenchmarking.orgMegahertz, More Is BetterPalabos 2.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3459.57 / Max: 3777Min: 2550 / Avg: 3430.77 / Max: 3841
OpenBenchmarking.orgWatts, Fewer Is BetterPalabos 2.3CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.01 / Avg: 255.67 / Max: 287.09Min: 33.19 / Avg: 266.48 / Max: 3101. 3DV Disabled: Approximate power consumption of 48918 Joules per run.2. Default: Approximate power consumption of 48454 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterPalabos 2.3CPU Temperature Monitor3DV DisabledDefault1326395265Min: 37.25 / Avg: 49.58 / Max: 63.88Min: 38.75 / Avg: 52.61 / Max: 66.25
OpenBenchmarking.orgMega Site Updates Per Second, More Is BetterPalabos 2.3Grid Size: 400Default3DV Disabled60120180240300Min: 313.81 / Avg: 317.82 / Max: 349.07Min: 289.6 / Avg: 290.19 / Max: 290.891. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm

GPAW

GPAW is a density-functional theory (DFT) Python code based on the projector-augmented wave (PAW) method and the atomic simulation environment (ASE). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 23.6Input: Carbon NanotubeDefault3DV Disabled918273645SE +/- 0.18, N = 3SE +/- 0.05, N = 334.7738.051. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi
OpenBenchmarking.orgMegahertz, More Is BetterGPAW 23.6CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 2975.6 / Max: 3732Min: 2550 / Avg: 2955.34 / Max: 3788
OpenBenchmarking.orgWatts, Fewer Is BetterGPAW 23.6CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.33 / Avg: 282.56 / Max: 326.44Min: 33.46 / Avg: 287.26 / Max: 331.31. 3DV Disabled: Approximate power consumption of 12056 Joules per run.2. Default: Approximate power consumption of 11490 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterGPAW 23.6CPU Temperature Monitor3DV DisabledDefault1326395265Min: 37.75 / Avg: 57.78 / Max: 65.88Min: 39.13 / Avg: 59.46 / Max: 67.38
OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 23.6Input: Carbon NanotubeDefault3DV Disabled816243240Min: 34.6 / Avg: 34.77 / Max: 35.13Min: 37.96 / Avg: 38.05 / Max: 38.121. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian Dragon ObjDefault3DV Disabled306090120150SE +/- 0.05, N = 4SE +/- 0.06, N = 4123.51113.24MIN: 121.61 / MAX: 126.87MIN: 111.56 / MAX: 116.79
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian Dragon ObjDefault3DV Disabled0.18110.36220.54330.72440.90550.8050.748
OpenBenchmarking.orgMegahertz, More Is BetterEmbree 4.1CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3225.4 / Max: 3699Min: 2550 / Avg: 3213.74 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 4.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.84 / Avg: 151.36 / Max: 309.88Min: 32.24 / Avg: 153.46 / Max: 328.591. 3DV Disabled: Approximate power consumption of 2800 Joules per run.2. Default: Approximate power consumption of 2801 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterEmbree 4.1CPU Temperature Monitor3DV DisabledDefault1020304050Min: 33.13 / Avg: 40 / Max: 48.25Min: 34.5 / Avg: 41.21 / Max: 50.63
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian Dragon ObjDefault3DV Disabled20406080100Min: 123.37 / Avg: 123.51 / Max: 123.62Min: 113.13 / Avg: 113.24 / Max: 113.41

libxsmm

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

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 256Default3DV Disabled7001400210028003500SE +/- 32.92, N = 5SE +/- 15.39, N = 33064.42811.71. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2
OpenBenchmarking.orgGFLOPS/s Per Watt, More Is Betterlibxsmm 2-1.17-3645M N K: 256Default3DV Disabled369121510.9849.557
OpenBenchmarking.orgMegahertz, More Is Betterlibxsmm 2-1.17-3645CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2449 / Avg: 3496.75 / Max: 3696Min: 2332 / Avg: 3335.15 / Max: 3688
OpenBenchmarking.orgWatts, Fewer Is Betterlibxsmm 2-1.17-3645CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 33.31 / Avg: 278.98 / Max: 358.89Min: 29.56 / Avg: 294.2 / Max: 338.481. Default: Approximate power consumption of 46645 Joules per run.2. 3DV Disabled: Approximate power consumption of 47660 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is Betterlibxsmm 2-1.17-3645CPU Temperature MonitorDefault3DV Disabled1326395265Min: 41.25 / Avg: 62.23 / Max: 65.38Min: 39.63 / Avg: 63.09 / Max: 65.5
OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 256Default3DV Disabled5001000150020002500Min: 2982 / Avg: 3064.44 / Max: 3126.7Min: 2787.4 / Avg: 2811.7 / Max: 2840.21. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -msse4.2

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian DragonDefault3DV Disabled306090120150SE +/- 0.10, N = 7SE +/- 0.10, N = 7144.00132.77MIN: 141.47 / MAX: 149.12MIN: 130.82 / MAX: 136.56
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian DragonDefault3DV Disabled0.17480.34960.52440.69920.8740.7770.739
OpenBenchmarking.orgMegahertz, More Is BetterEmbree 4.1CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2796.13 / Max: 3695Min: 2550 / Avg: 2784.91 / Max: 3705
OpenBenchmarking.orgWatts, Fewer Is BetterEmbree 4.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.71 / Avg: 179.74 / Max: 311.66Min: 32.63 / Avg: 185.23 / Max: 331.221. 3DV Disabled: Approximate power consumption of 1695 Joules per run.2. Default: Approximate power consumption of 1667 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterEmbree 4.1CPU Temperature Monitor3DV DisabledDefault1122334455Min: 34.75 / Avg: 43.16 / Max: 50.25Min: 36 / Avg: 44.19 / Max: 53.38
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian DragonDefault3DV Disabled306090120150Min: 143.61 / Avg: 144 / Max: 144.37Min: 132.28 / Avg: 132.77 / Max: 133.1

LULESH

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

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3Default3DV Disabled7K14K21K28K35KSE +/- 182.94, N = 4SE +/- 202.03, N = 430715.3328515.361. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi
OpenBenchmarking.orgz/s Per Watt, More Is BetterLULESH 2.0.3Default3DV Disabled4080120160200172.34163.81
OpenBenchmarking.orgMegahertz, More Is BetterLULESH 2.0.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3368.1 / Max: 3704Min: 2550 / Avg: 3338.45 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterLULESH 2.0.3CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 29.32 / Avg: 174.08 / Max: 274.85Min: 33.27 / Avg: 178.23 / Max: 283.511. 3DV Disabled: Approximate power consumption of 3569 Joules per run.2. Default: Approximate power consumption of 3431 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterLULESH 2.0.3CPU Temperature Monitor3DV DisabledDefault1428425670Min: 38.25 / Avg: 47.44 / Max: 62.5Min: 39.13 / Avg: 48.86 / Max: 72
OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3Default3DV Disabled5K10K15K20K25KMin: 30272.24 / Avg: 30715.33 / Max: 31134.4Min: 27931.34 / Avg: 28515.36 / Max: 28861.661. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - DegriddingDefault3DV Disabled3K6K9K12K15KSE +/- 16.78, N = 3SE +/- 45.70, N = 315603.214487.71. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMillion Grid Points Per Second Per Watt, More Is BetterASKAP 1.0Test: tConvolve MT - DegriddingDefault3DV Disabled20406080100107.43102.92
OpenBenchmarking.orgMegahertz, More Is BetterASKAP 1.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3614.57 / Max: 3727Min: 2550 / Avg: 3607.86 / Max: 3717
OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 29.57 / Avg: 140.76 / Max: 281.09Min: 33.19 / Avg: 145.24 / Max: 297.751. 3DV Disabled: Approximate power consumption of 43214 Joules per run.2. Default: Approximate power consumption of 43813 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterASKAP 1.0CPU Temperature Monitor3DV DisabledDefault1122334455Min: 34.25 / Avg: 38.26 / Max: 50.75Min: 34.25 / Avg: 38.78 / Max: 54.5
OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - DegriddingDefault3DV Disabled3K6K9K12K15KMin: 15571.5 / Avg: 15603.2 / Max: 15628.6Min: 14396.3 / Avg: 14487.67 / Max: 14535.41. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

PyHPC Benchmarks

PyHPC-Benchmarks is a suite of Python high performance computing benchmarks for execution on CPUs and GPUs using various popular Python HPC libraries. The PyHPC CPU-based benchmarks focus on sequential CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPyHPC Benchmarks 3.0Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral MixingDefault3DV Disabled0.38160.76321.14481.52641.908SE +/- 0.001, N = 3SE +/- 0.005, N = 31.5781.696
OpenBenchmarking.orgMegahertz, More Is BetterPyHPC Benchmarks 3.0CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3652.89 / Max: 3719Min: 2550 / Avg: 3648.07 / Max: 3730
OpenBenchmarking.orgWatts, Fewer Is BetterPyHPC Benchmarks 3.0CPU Power Consumption Monitor3DV DisabledDefault20406080100Min: 27.95 / Avg: 102.83 / Max: 116.37Min: 16.16 / Avg: 106.96 / Max: 121.371. 3DV Disabled: Approximate power consumption of 9357 Joules per run.2. Default: Approximate power consumption of 9341 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterPyHPC Benchmarks 3.0CPU Temperature MonitorDefault3DV Disabled1122334455Min: 30.38 / Avg: 36.23 / Max: 47.38Min: 28.5 / Avg: 42.74 / Max: 53.5
OpenBenchmarking.orgSeconds, Fewer Is BetterPyHPC Benchmarks 3.0Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral MixingDefault3DV Disabled246810Min: 1.58 / Avg: 1.58 / Max: 1.58Min: 1.69 / Avg: 1.7 / Max: 1.7

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: PipeDefault3DV Disabled13M26M39M52M65MSE +/- 612854.45, N = 3SE +/- 848784.40, N = 1560302912.6756210994.921. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: PipeDefault3DV Disabled40K80K120K160K200K208554.90198855.26
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2952.25 / Max: 3014Min: 2550 / Avg: 2930.57 / Max: 3694
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.31 / Avg: 282.67 / Max: 319.86Min: 32.77 / Avg: 289.15 / Max: 326.221. 3DV Disabled: Approximate power consumption of 9592 Joules per run.2. Default: Approximate power consumption of 9831 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature MonitorDefault3DV Disabled1326395265Min: 38.75 / Avg: 58.59 / Max: 63.38Min: 39.13 / Avg: 59.94 / Max: 65.75
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: PipeDefault3DV Disabled10M20M30M40M50MMin: 59322292.85 / Avg: 60302912.67 / Max: 61430058.74Min: 50414851.57 / Avg: 56210994.92 / Max: 61314920.821. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: FutexDefault3DV Disabled900K1800K2700K3600K4500KSE +/- 13786.79, N = 3SE +/- 37571.61, N = 33985709.563716134.141. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: FutexDefault3DV Disabled5K10K15K20K25K22345.7621473.84
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3397.54 / Max: 3699Min: 2550 / Avg: 3365.24 / Max: 3694
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault4080120160200Min: 29.26 / Avg: 173.05 / Max: 201.38Min: 32.34 / Avg: 178.37 / Max: 199.051. 3DV Disabled: Approximate power consumption of 5942 Joules per run.2. Default: Approximate power consumption of 6064 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature MonitorDefault3DV Disabled1122334455Min: 37 / Avg: 44.34 / Max: 46.5Min: 37.38 / Avg: 44.46 / Max: 52.63
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: FutexDefault3DV Disabled700K1400K2100K2800K3500KMin: 3960880.75 / Avg: 3985709.56 / Max: 4008510.23Min: 3678150.46 / Avg: 3716134.14 / Max: 3791275.851. (CXX) g++ options: -O2 -std=gnu99 -lc

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - GriddingDefault3DV Disabled3K6K9K12K15KSE +/- 1.20, N = 3SE +/- 1.82, N = 313582.812713.51. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - GriddingDefault3DV Disabled2K4K6K8K10KMin: 13581.6 / Avg: 13582.8 / Max: 13585.2Min: 12710.4 / Avg: 12713.53 / Max: 12716.71. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Ngspice

Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552Default3DV Disabled20406080100SE +/- 0.08, N = 3SE +/- 0.13, N = 3100.85107.581. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE
OpenBenchmarking.orgMegahertz, More Is BetterNgspice 34CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3667.27 / Max: 3740Min: 2550 / Avg: 3658.42 / Max: 3710
OpenBenchmarking.orgWatts, Fewer Is BetterNgspice 34CPU Power Consumption Monitor3DV DisabledDefault20406080100Min: 27.65 / Avg: 99.12 / Max: 103.53Min: 31.83 / Avg: 103.51 / Max: 108.481. 3DV Disabled: Approximate power consumption of 11432 Joules per run.2. Default: Approximate power consumption of 11283 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNgspice 34CPU Temperature Monitor3DV DisabledDefault918273645Min: 27.63 / Avg: 33.02 / Max: 33.88Min: 28.63 / Avg: 34.85 / Max: 42.25
OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552Default3DV Disabled20406080100Min: 100.71 / Avg: 100.85 / Max: 100.98Min: 107.34 / Avg: 107.58 / Max: 107.771. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: ExhaustiveDefault3DV Disabled246810SE +/- 0.0017, N = 4SE +/- 0.0018, N = 46.14115.75981. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.0Preset: ExhaustiveDefault3DV Disabled0.0070.0140.0210.0280.0350.0310.030
OpenBenchmarking.orgMegahertz, More Is BetterASTC Encoder 4.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3030.7 / Max: 3699Min: 2550 / Avg: 2929.86 / Max: 3700
OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.38 / Avg: 190.96 / Max: 309.33Min: 31.79 / Avg: 201.1 / Max: 335.351. 3DV Disabled: Approximate power consumption of 3103 Joules per run.2. Default: Approximate power consumption of 3167 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterASTC Encoder 4.0CPU Temperature Monitor3DV DisabledDefault1122334455Min: 30.25 / Avg: 43.29 / Max: 52.5Min: 30.75 / Avg: 44.1 / Max: 54.75
OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: ExhaustiveDefault3DV Disabled246810Min: 6.14 / Avg: 6.14 / Max: 6.14Min: 5.76 / Avg: 5.76 / Max: 5.761. (CXX) g++ options: -O3 -flto -pthread

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Fused Multiply-AddDefault3DV Disabled16M32M48M64M80MSE +/- 17500.16, N = 3SE +/- 33044.58, N = 376566577.1271844167.061. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Fused Multiply-Add3DV DisabledDefault60K120K180K240K300K279828.29270844.71
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2785.36 / Max: 3691Min: 2550 / Avg: 2633.96 / Max: 3691
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 31.09 / Avg: 256.74 / Max: 289.19Min: 34 / Avg: 282.7 / Max: 318.381. 3DV Disabled: Approximate power consumption of 8644 Joules per run.2. Default: Approximate power consumption of 9612 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1224364860Min: 38.75 / Avg: 55.49 / Max: 59.25Min: 36.38 / Avg: 56.38 / Max: 61.25
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Fused Multiply-AddDefault3DV Disabled13M26M39M52M65MMin: 76545144.47 / Avg: 76566577.12 / Max: 76601256.91Min: 71787593.87 / Avg: 71844167.06 / Max: 71902041.481. (CXX) g++ options: -O2 -std=gnu99 -lc

Palabos

The Palabos library is a framework for general purpose Computational Fluid Dynamics (CFD). Palabos uses a kernel based on the Lattice Boltzmann method. This test profile uses the Palabos MPI-based Cavity3D benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMega Site Updates Per Second, More Is BetterPalabos 2.3Grid Size: 500Default3DV Disabled70140210280350SE +/- 0.10, N = 3SE +/- 0.57, N = 3328.71308.471. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm
OpenBenchmarking.orgMega Site Updates Per Second Per Watt, More Is BetterPalabos 2.3Grid Size: 500Default3DV Disabled0.28370.56740.85111.13481.41851.2611.240
OpenBenchmarking.orgMegahertz, More Is BetterPalabos 2.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3452.59 / Max: 3712Min: 2550 / Avg: 3439.43 / Max: 3727
OpenBenchmarking.orgWatts, Fewer Is BetterPalabos 2.3CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 28.97 / Avg: 248.67 / Max: 304.71Min: 32.52 / Avg: 260.6 / Max: 303.411. 3DV Disabled: Approximate power consumption of 46585 Joules per run.2. Default: Approximate power consumption of 45779 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterPalabos 2.3CPU Temperature Monitor3DV DisabledDefault1224364860Min: 34.13 / Avg: 49.43 / Max: 60.88Min: 36 / Avg: 51.12 / Max: 62
OpenBenchmarking.orgMega Site Updates Per Second, More Is BetterPalabos 2.3Grid Size: 500Default3DV Disabled60120180240300Min: 328.6 / Avg: 328.71 / Max: 328.92Min: 307.72 / Avg: 308.47 / Max: 309.61. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm

miniFE

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

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallDefault3DV Disabled12K24K36K48K60KSE +/- 227.90, N = 5SE +/- 56.96, N = 554408.251223.51. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi
OpenBenchmarking.orgCG Mflops Per Watt, More Is BetterminiFE 2.2Problem Size: SmallDefault3DV Disabled80160240320400349.97338.97
OpenBenchmarking.orgMegahertz, More Is BetterminiFE 2.2CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3220.03 / Max: 3700Min: 2550 / Avg: 3176.49 / Max: 3698
OpenBenchmarking.orgWatts, Fewer Is BetterminiFE 2.2CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 28.76 / Avg: 151.12 / Max: 288.84Min: 32.98 / Avg: 155.47 / Max: 304.051. 3DV Disabled: Approximate power consumption of 1995 Joules per run.2. Default: Approximate power consumption of 2021 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterminiFE 2.2CPU Temperature Monitor3DV DisabledDefault1122334455Min: 35.63 / Avg: 41.72 / Max: 51.25Min: 36.88 / Avg: 42.98 / Max: 52.63
OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallDefault3DV Disabled9K18K27K36K45KMin: 54064.7 / Avg: 54408.18 / Max: 55241.7Min: 51054.8 / Avg: 51223.46 / Max: 51370.71. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi

Timed Linux Kernel Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: allmodconfigDefault3DV Disabled50100150200250SE +/- 0.58, N = 3SE +/- 0.76, N = 3202.56215.05
OpenBenchmarking.orgMegahertz, More Is BetterTimed Linux Kernel Compilation 6.1CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3257.49 / Max: 3712Min: 2550 / Avg: 3213.71 / Max: 3745
OpenBenchmarking.orgWatts, Fewer Is BetterTimed Linux Kernel Compilation 6.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.24 / Avg: 259.88 / Max: 310.48Min: 32.48 / Avg: 272.39 / Max: 327.241. 3DV Disabled: Approximate power consumption of 58474 Joules per run.2. Default: Approximate power consumption of 57748 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterTimed Linux Kernel Compilation 6.1CPU Temperature Monitor3DV DisabledDefault1428425670Min: 32.38 / Avg: 58.09 / Max: 72.5Min: 34.13 / Avg: 59.71 / Max: 74.88
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: allmodconfigDefault3DV Disabled4080120160200Min: 201.49 / Avg: 202.56 / Max: 203.5Min: 213.92 / Avg: 215.05 / Max: 216.5

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamDefault3DV Disabled306090120150SE +/- 0.07, N = 3SE +/- 0.07, N = 3145.05153.87
OpenBenchmarking.orgMegahertz, More Is BetterNeural Magic DeepSparse 1.5CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3085.77 / Max: 3712Min: 2550 / Avg: 3085.01 / Max: 3731
OpenBenchmarking.orgWatts, Fewer Is BetterNeural Magic DeepSparse 1.5CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29 / Avg: 250.11 / Max: 317.34Min: 33.18 / Avg: 262.66 / Max: 321.661. 3DV Disabled: Approximate power consumption of 12505 Joules per run.2. Default: Approximate power consumption of 12695 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNeural Magic DeepSparse 1.5CPU Temperature Monitor3DV DisabledDefault1224364860Min: 38.75 / Avg: 54.54 / Max: 61.63Min: 38.25 / Avg: 54.91 / Max: 61.75
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamDefault3DV Disabled306090120150Min: 144.95 / Avg: 145.05 / Max: 145.19Min: 153.74 / Avg: 153.87 / Max: 153.97

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: ThoroughDefault3DV Disabled1326395265SE +/- 0.02, N = 6SE +/- 0.01, N = 556.7853.541. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.0Preset: ThoroughDefault3DV Disabled0.07790.15580.23370.31160.38950.3460.345
OpenBenchmarking.orgMegahertz, More Is BetterASTC Encoder 4.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3121.27 / Max: 3705Min: 2550 / Avg: 3020.15 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 27.77 / Avg: 155.41 / Max: 309.08Min: 31.78 / Avg: 164.25 / Max: 335.571. 3DV Disabled: Approximate power consumption of 1927 Joules per run.2. Default: Approximate power consumption of 1998 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterASTC Encoder 4.0CPU Temperature Monitor3DV DisabledDefault1122334455Min: 28.38 / Avg: 38.67 / Max: 49.25Min: 28.5 / Avg: 39.8 / Max: 52.5
OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: ThoroughDefault3DV Disabled1122334455Min: 56.71 / Avg: 56.78 / Max: 56.87Min: 53.51 / Avg: 53.54 / Max: 53.561. (CXX) g++ options: -O3 -flto -pthread

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamDefault3DV Disabled70140210280350SE +/- 0.25, N = 3SE +/- 0.17, N = 3329.98311.20
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamDefault3DV Disabled60120180240300Min: 329.48 / Avg: 329.98 / Max: 330.25Min: 310.9 / Avg: 311.2 / Max: 311.48

ACES DGEMM

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

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateDefault3DV Disabled918273645SE +/- 0.13, N = 7SE +/- 0.32, N = 840.5538.261. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterACES DGEMM 1.0Sustained Floating-Point Rate3DV DisabledDefault0.05220.10440.15660.20880.2610.2320.229
OpenBenchmarking.orgMegahertz, More Is BetterACES DGEMM 1.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2933.93 / Max: 3694Min: 2550 / Avg: 2910.32 / Max: 3694
OpenBenchmarking.orgWatts, Fewer Is BetterACES DGEMM 1.0CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 14.69 / Avg: 165.07 / Max: 273.78Min: 32.16 / Avg: 176.93 / Max: 299.71. 3DV Disabled: Approximate power consumption of 1651 Joules per run.2. Default: Approximate power consumption of 1744 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterACES DGEMM 1.0CPU Temperature Monitor3DV DisabledDefault1020304050Min: 32.25 / Avg: 41.19 / Max: 48.88Min: 33.13 / Avg: 42.81 / Max: 52.13
OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateDefault3DV Disabled816243240Min: 39.88 / Avg: 40.55 / Max: 40.95Min: 37.11 / Avg: 38.26 / Max: 39.351. (CC) gcc options: -O3 -march=native -fopenmp

Timed Godot Game Engine Compilation

This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To CompileDefault3DV Disabled20406080100SE +/- 0.06, N = 3SE +/- 0.11, N = 388.3893.64
OpenBenchmarking.orgMegahertz, More Is BetterTimed Godot Game Engine Compilation 4.0CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3402.96 / Max: 3717Min: 2550 / Avg: 3388.92 / Max: 3742
OpenBenchmarking.orgWatts, Fewer Is BetterTimed Godot Game Engine Compilation 4.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.21 / Avg: 182.1 / Max: 322.23Min: 32.03 / Avg: 189.61 / Max: 334.811. 3DV Disabled: Approximate power consumption of 18089 Joules per run.2. Default: Approximate power consumption of 17697 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterTimed Godot Game Engine Compilation 4.0CPU Temperature Monitor3DV DisabledDefault1326395265Min: 29.5 / Avg: 46.29 / Max: 64.13Min: 31.25 / Avg: 48.07 / Max: 65.13
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To CompileDefault3DV Disabled20406080100Min: 88.3 / Avg: 88.38 / Max: 88.51Min: 93.48 / Avg: 93.64 / Max: 93.84

Timed LLVM Compilation

This test times how long it takes to compile/build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaDefault3DV Disabled306090120150SE +/- 0.31, N = 3SE +/- 0.24, N = 3112.89119.45
OpenBenchmarking.orgMegahertz, More Is BetterTimed LLVM Compilation 16.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 3207.73 / Max: 3738Min: 2550 / Avg: 3207.45 / Max: 3754
OpenBenchmarking.orgWatts, Fewer Is BetterTimed LLVM Compilation 16.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.73 / Avg: 229.21 / Max: 319.56Min: 32.56 / Avg: 238.86 / Max: 331.711. 3DV Disabled: Approximate power consumption of 30331 Joules per run.2. Default: Approximate power consumption of 30176 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterTimed LLVM Compilation 16.0CPU Temperature Monitor3DV DisabledDefault1326395265Min: 36 / Avg: 54.05 / Max: 69.13Min: 37.38 / Avg: 55.73 / Max: 68.75
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaDefault3DV Disabled20406080100Min: 112.39 / Avg: 112.89 / Max: 113.45Min: 119.03 / Avg: 119.45 / Max: 119.84

Palabos

The Palabos library is a framework for general purpose Computational Fluid Dynamics (CFD). Palabos uses a kernel based on the Lattice Boltzmann method. This test profile uses the Palabos MPI-based Cavity3D benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMega Site Updates Per Second, More Is BetterPalabos 2.3Grid Size: 1000Default3DV Disabled80160240320400SE +/- 0.10, N = 3SE +/- 0.97, N = 3370.19350.091. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm
OpenBenchmarking.orgMega Site Updates Per Second Per Watt, More Is BetterPalabos 2.3Grid Size: 1000Default3DV Disabled0.3330.6660.9991.3321.6651.4801.463
OpenBenchmarking.orgMegahertz, More Is BetterPalabos 2.3CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 3443.31 / Max: 3700Min: 2550 / Avg: 3441.06 / Max: 3838
OpenBenchmarking.orgWatts, Fewer Is BetterPalabos 2.3CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.03 / Avg: 239.28 / Max: 314.49Min: 32.8 / Avg: 250.19 / Max: 319.421. 3DV Disabled: Approximate power consumption of 42911 Joules per run.2. Default: Approximate power consumption of 42699 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterPalabos 2.3CPU Temperature Monitor3DV DisabledDefault1224364860Min: 33.5 / Avg: 47.62 / Max: 62.75Min: 35 / Avg: 49.49 / Max: 62.75
OpenBenchmarking.orgMega Site Updates Per Second, More Is BetterPalabos 2.3Grid Size: 1000Default3DV Disabled70140210280350Min: 370.07 / Avg: 370.19 / Max: 370.39Min: 348.55 / Avg: 350.09 / Max: 351.891. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerDefault3DV Disabled2004006008001000SE +/- 2.33, N = 3SE +/- 1.33, N = 3106511261. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2913.9 / Max: 3703Min: 2550 / Avg: 2832.45 / Max: 3700
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.34 / Avg: 275.4 / Max: 311.11Min: 34.62 / Avg: 293.15 / Max: 331.681. 3DV Disabled: Approximate power consumption of 31855 Joules per run.2. Default: Approximate power consumption of 33810 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1326395265Min: 40.13 / Avg: 56.77 / Max: 61.63Min: 42 / Avg: 60.34 / Max: 65.38
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerDefault3DV Disabled2004006008001000Min: 1063 / Avg: 1065.33 / Max: 1070Min: 1123 / Avg: 1125.67 / Max: 11271. (CXX) g++ options: -O3 -lm -ldl

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector Floating PointDefault3DV Disabled60K120K180K240K300KSE +/- 151.27, N = 3SE +/- 430.31, N = 3257925.61244276.801. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Vector Floating Point3DV DisabledDefault2004006008001000944.48914.27
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2955.25 / Max: 3429Min: 2550 / Avg: 2867.02 / Max: 3723
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 31.1 / Avg: 258.64 / Max: 294.01Min: 34.46 / Avg: 282.11 / Max: 320.241. 3DV Disabled: Approximate power consumption of 8880 Joules per run.2. Default: Approximate power consumption of 9592 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1224364860Min: 38.25 / Avg: 55.65 / Max: 60.38Min: 38.25 / Avg: 57.62 / Max: 63.63
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector Floating PointDefault3DV Disabled40K80K120K160K200KMin: 257658.71 / Avg: 257925.61 / Max: 258182.44Min: 243540.69 / Avg: 244276.8 / Max: 2450311. (CXX) g++ options: -O2 -std=gnu99 -lc

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerDefault3DV Disabled4K8K12K16K20KSE +/- 6.03, N = 3SE +/- 28.09, N = 317078180301. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 2919.77 / Max: 3703Min: 2550 / Avg: 2847.57 / Max: 3772
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.37 / Avg: 274.15 / Max: 312.07Min: 33.72 / Avg: 288.79 / Max: 331.261. 3DV Disabled: Approximate power consumption of 29243 Joules per run.2. Default: Approximate power consumption of 29456 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1326395265Min: 40.63 / Avg: 56.34 / Max: 61.13Min: 42.38 / Avg: 59.33 / Max: 64.63
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerDefault3DV Disabled3K6K9K12K15KMin: 17071 / Avg: 17078 / Max: 17090Min: 17974 / Avg: 18029.67 / Max: 180641. (CXX) g++ options: -O3 -lm -ldl

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerDefault3DV Disabled4K8K12K16K20KSE +/- 50.26, N = 3SE +/- 41.97, N = 316953178791. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2919.36 / Max: 3703Min: 2550 / Avg: 2838.62 / Max: 3707
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.99 / Avg: 274.5 / Max: 312.77Min: 33.99 / Avg: 289.47 / Max: 333.121. 3DV Disabled: Approximate power consumption of 29006 Joules per run.2. Default: Approximate power consumption of 29236 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1428425670Min: 40.38 / Avg: 57.13 / Max: 66.88Min: 42.25 / Avg: 60.21 / Max: 72.88
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerDefault3DV Disabled3K6K9K12K15KMin: 16864 / Avg: 16952.67 / Max: 17038Min: 17817 / Avg: 17879 / Max: 179591. (CXX) g++ options: -O3 -lm -ldl

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerDefault3DV Disabled30060090012001500SE +/- 0.33, N = 3SE +/- 1.53, N = 3126113291. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2891.16 / Max: 3705Min: 2550 / Avg: 2831.94 / Max: 3706
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.93 / Avg: 277.42 / Max: 312.57Min: 35.36 / Avg: 296.21 / Max: 332.241. 3DV Disabled: Approximate power consumption of 32643 Joules per run.2. Default: Approximate power consumption of 34756 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1326395265Min: 40.38 / Avg: 56.97 / Max: 62Min: 42.88 / Avg: 60.74 / Max: 69.38
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerDefault3DV Disabled2004006008001000Min: 1260 / Avg: 1260.67 / Max: 1261Min: 1326 / Avg: 1329 / Max: 13311. (CXX) g++ options: -O3 -lm -ldl

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: LionDefault3DV Disabled11002200330044005500SE +/- 2.98, N = 7SE +/- 18.65, N = 6501152781. (CXX) g++ options: -O3
OpenBenchmarking.orgMegahertz, More Is BetterGoogle Draco 1.5.6CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3361.86 / Max: 3705Min: 2550 / Avg: 3349.19 / Max: 3700
OpenBenchmarking.orgWatts, Fewer Is BetterGoogle Draco 1.5.6CPU Power Consumption Monitor3DV DisabledDefault20406080100Min: 28.7 / Avg: 77.67 / Max: 105.66Min: 32.25 / Avg: 79.38 / Max: 109.191. 3DV Disabled: Approximate power consumption of 841 Joules per run.2. Default: Approximate power consumption of 816 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterGoogle Draco 1.5.6CPU Temperature MonitorDefault3DV Disabled918273645Min: 31.88 / Avg: 37.44 / Max: 42.25Min: 32.88 / Avg: 37.96 / Max: 42.63
OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: LionDefault3DV Disabled9001800270036004500Min: 4996 / Avg: 5010.57 / Max: 5018Min: 5209 / Avg: 5278.17 / Max: 53511. (CXX) g++ options: -O3

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerDefault3DV Disabled9K18K27K36K45KSE +/- 43.02, N = 3SE +/- 38.63, N = 340392425241. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2876.88 / Max: 3701Min: 2550 / Avg: 2811.01 / Max: 3699
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.96 / Avg: 284.39 / Max: 313.62Min: 34.04 / Avg: 301.23 / Max: 333.341. 3DV Disabled: Approximate power consumption of 40762 Joules per run.2. Default: Approximate power consumption of 41268 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1326395265Min: 40.13 / Avg: 57.69 / Max: 61.63Min: 41.25 / Avg: 60.59 / Max: 65.38
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerDefault3DV Disabled7K14K21K28K35KMin: 40323 / Avg: 40392 / Max: 40471Min: 42448 / Avg: 42524 / Max: 425741. (CXX) g++ options: -O3 -lm -ldl

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerDefault3DV Disabled2004006008001000SE +/- 1.15, N = 3SE +/- 0.67, N = 3105911141. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2903.46 / Max: 3697Min: 2521 / Avg: 2818.68 / Max: 3726
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.94 / Avg: 276.07 / Max: 311.59Min: 33.17 / Avg: 292.75 / Max: 331.51. 3DV Disabled: Approximate power consumption of 31749 Joules per run.2. Default: Approximate power consumption of 33373 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1326395265Min: 30.38 / Avg: 54.91 / Max: 61.13Min: 31.25 / Avg: 58.18 / Max: 65
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path TracerDefault3DV Disabled2004006008001000Min: 1057 / Avg: 1059 / Max: 1061Min: 1113 / Avg: 1114.33 / Max: 11151. (CXX) g++ options: -O3 -lm -ldl

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerDefault3DV Disabled8K16K24K32K40KSE +/- 51.48, N = 3SE +/- 46.77, N = 334341360731. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2891.46 / Max: 3700Min: 2550 / Avg: 2818.86 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.21 / Avg: 279 / Max: 311.98Min: 33.36 / Avg: 295.06 / Max: 332.31. 3DV Disabled: Approximate power consumption of 34596 Joules per run.2. Default: Approximate power consumption of 35013 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1428425670Min: 40.13 / Avg: 57.65 / Max: 63.13Min: 41.5 / Avg: 60 / Max: 72.88
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerDefault3DV Disabled6K12K18K24K30KMin: 34242 / Avg: 34341 / Max: 34415Min: 35991 / Avg: 36072.67 / Max: 361531. (CXX) g++ options: -O3 -lm -ldl

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Barbershop - Compute: CPU-OnlyDefault3DV Disabled306090120150SE +/- 0.01, N = 3SE +/- 0.33, N = 3142.03149.19
OpenBenchmarking.orgMegahertz, More Is BetterBlender 3.6CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2793.62 / Max: 3698Min: 2550 / Avg: 2720.1 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.6CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.16 / Avg: 305.66 / Max: 322.26Min: 33.97 / Avg: 318.93 / Max: 337.161. 3DV Disabled: Approximate power consumption of 46970 Joules per run.2. Default: Approximate power consumption of 46776 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterBlender 3.6CPU Temperature Monitor3DV DisabledDefault1326395265Min: 37.75 / Avg: 61.2 / Max: 66.88Min: 40.13 / Avg: 64.06 / Max: 67.75
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Barbershop - Compute: CPU-OnlyDefault3DV Disabled306090120150Min: 142.02 / Avg: 142.03 / Max: 142.04Min: 148.69 / Avg: 149.19 / Max: 149.8

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/ao/real_timeDefault3DV Disabled612182430SE +/- 0.12, N = 3SE +/- 0.06, N = 326.7925.52
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/ao/real_time3DV DisabledDefault0.02270.04540.06810.09080.11350.1010.098
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay 2.12CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3044.11 / Max: 3704Min: 2550 / Avg: 2909.46 / Max: 3698
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.12CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.21 / Avg: 252.27 / Max: 289.84Min: 33.45 / Avg: 272.37 / Max: 314.911. 3DV Disabled: Approximate power consumption of 13875 Joules per run.2. Default: Approximate power consumption of 15071 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay 2.12CPU Temperature Monitor3DV DisabledDefault1224364860Min: 35 / Avg: 53.11 / Max: 58.5Min: 36.38 / Avg: 56.4 / Max: 62.75
OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/ao/real_timeDefault3DV Disabled612182430Min: 26.6 / Avg: 26.79 / Max: 27Min: 25.41 / Avg: 25.52 / Max: 25.6

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeDefault3DV Disabled612182430SE +/- 0.04, N = 3SE +/- 0.07, N = 325.9324.70
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/scivis/real_time3DV DisabledDefault0.02180.04360.06540.08720.1090.0970.094
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay 2.12CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3037.3 / Max: 3698Min: 2550 / Avg: 2933.47 / Max: 3698
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.12CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.95 / Avg: 253.76 / Max: 292.25Min: 33.47 / Avg: 275.34 / Max: 317.161. 3DV Disabled: Approximate power consumption of 14126 Joules per run.2. Default: Approximate power consumption of 15328 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay 2.12CPU Temperature Monitor3DV DisabledDefault1224364860Min: 37.75 / Avg: 54.12 / Max: 59.25Min: 40.13 / Avg: 57.99 / Max: 63.88
OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeDefault3DV Disabled612182430Min: 25.85 / Avg: 25.93 / Max: 25.97Min: 24.62 / Avg: 24.7 / Max: 24.84

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerDefault3DV Disabled5K10K15K20K25KSE +/- 50.67, N = 3SE +/- 17.14, N = 320223212221. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2905.18 / Max: 3699Min: 2550 / Avg: 2840.65 / Max: 3705
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.5 / Avg: 272.17 / Max: 313.28Min: 33.8 / Avg: 289.71 / Max: 332.631. 3DV Disabled: Approximate power consumption of 27670 Joules per run.2. Default: Approximate power consumption of 28198 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1326395265Min: 40.13 / Avg: 56.14 / Max: 61.13Min: 42 / Avg: 58.93 / Max: 64.13
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path TracerDefault3DV Disabled4K8K12K16K20KMin: 20122 / Avg: 20223.33 / Max: 20275Min: 21189 / Avg: 21221.67 / Max: 212471. (CXX) g++ options: -O3 -lm -ldl

HeFFTe - Highly Efficient FFT for Exascale

HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 512Default3DV Disabled1530456075SE +/- 0.10, N = 3SE +/- 0.01, N = 367.8564.701. (CXX) g++ options: -O3
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 512Default3DV Disabled0.0610.1220.1830.2440.3050.2710.270
OpenBenchmarking.orgMegahertz, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3388.89 / Max: 3702Min: 2550 / Avg: 3341.26 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 28.78 / Avg: 239.78 / Max: 279.22Min: 31.94 / Avg: 250.53 / Max: 302.431. 3DV Disabled: Approximate power consumption of 8233 Joules per run.2. Default: Approximate power consumption of 8184 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3CPU Temperature Monitor3DV DisabledDefault1224364860Min: 29.38 / Avg: 46.81 / Max: 60.25Min: 30.25 / Avg: 47.85 / Max: 59.38
OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 512Default3DV Disabled1326395265Min: 67.66 / Avg: 67.85 / Max: 67.95Min: 64.69 / Avg: 64.7 / Max: 64.711. (CXX) g++ options: -O3

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerDefault3DV Disabled8K16K24K32K40KSE +/- 22.67, N = 3SE +/- 22.19, N = 334007356611. (CXX) g++ options: -O3 -lm -ldl
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay Studio 0.11CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2912.24 / Max: 3706Min: 2550 / Avg: 2814.74 / Max: 3700
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay Studio 0.11CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 30.23 / Avg: 279.4 / Max: 312.94Min: 33.28 / Avg: 295.38 / Max: 332.831. 3DV Disabled: Approximate power consumption of 34366 Joules per run.2. Default: Approximate power consumption of 34855 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay Studio 0.11CPU Temperature Monitor3DV DisabledDefault1326395265Min: 40.63 / Avg: 57.37 / Max: 66.25Min: 41.5 / Avg: 60.29 / Max: 65
OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 0.11Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path TracerDefault3DV Disabled6K12K18K24K30KMin: 33962 / Avg: 34007.33 / Max: 34031Min: 35626 / Avg: 35660.67 / Max: 357021. (CXX) g++ options: -O3 -lm -ldl

NAMD

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

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsDefault3DV Disabled0.05830.11660.17490.23320.2915SE +/- 0.00040, N = 3SE +/- 0.00025, N = 40.247330.25909
OpenBenchmarking.orgMegahertz, More Is BetterNAMD 2.14CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2954.12 / Max: 3699Min: 2517 / Avg: 2833.73 / Max: 3696
OpenBenchmarking.orgWatts, Fewer Is BetterNAMD 2.14CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.91 / Avg: 253.28 / Max: 318.62Min: 32.81 / Avg: 292.64 / Max: 335.191. 3DV Disabled: Approximate power consumption of 5129 Joules per run.2. Default: Approximate power consumption of 9560 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNAMD 2.14CPU Temperature Monitor3DV DisabledDefault1224364860Min: 33.63 / Avg: 51.91 / Max: 59.25Min: 34.5 / Avg: 57.66 / Max: 63.63
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsDefault3DV Disabled12345Min: 0.25 / Avg: 0.25 / Max: 0.25Min: 0.26 / Avg: 0.26 / Max: 0.26

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 19.8.1Time To CompileDefault3DV Disabled20406080100SE +/- 0.15, N = 3SE +/- 0.35, N = 3105.23110.20
OpenBenchmarking.orgMegahertz, More Is BetterTimed Node.js Compilation 19.8.1CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3303.81 / Max: 3761Min: 2550 / Avg: 3293.39 / Max: 3729
OpenBenchmarking.orgWatts, Fewer Is BetterTimed Node.js Compilation 19.8.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.63 / Avg: 218.1 / Max: 328.35Min: 33.32 / Avg: 225.5 / Max: 338.661. 3DV Disabled: Approximate power consumption of 25590 Joules per run.2. Default: Approximate power consumption of 25406 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterTimed Node.js Compilation 19.8.1CPU Temperature Monitor3DV DisabledDefault1326395265Min: 34.13 / Avg: 51.63 / Max: 68.38Min: 35.38 / Avg: 53.56 / Max: 67.25
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 19.8.1Time To CompileDefault3DV Disabled20406080100Min: 104.92 / Avg: 105.23 / Max: 105.39Min: 109.55 / Avg: 110.2 / Max: 110.76

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Memory CopyingDefault3DV Disabled7K14K21K28K35KSE +/- 1.22, N = 3SE +/- 4.41, N = 332994.1831510.801. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Memory Copying3DV DisabledDefault306090120150127.63121.73
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3035.88 / Max: 3118Min: 2550 / Avg: 2945.73 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.99 / Avg: 246.89 / Max: 278.45Min: 34.96 / Avg: 271.05 / Max: 307.131. 3DV Disabled: Approximate power consumption of 8394 Joules per run.2. Default: Approximate power consumption of 9216 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1326395265Min: 34.5 / Avg: 55.74 / Max: 61.25Min: 38.25 / Avg: 60 / Max: 64.88
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Memory CopyingDefault3DV Disabled6K12K18K24K30KMin: 32992.45 / Avg: 32994.18 / Max: 32996.54Min: 31502.12 / Avg: 31510.8 / Max: 31516.461. (CXX) g++ options: -O2 -std=gnu99 -lc

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 144 144 144 - RT: 60Default3DV Disabled612182430SE +/- 0.44, N = 9SE +/- 0.21, N = 324.6123.521. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 144 144 144 - RT: 60Default3DV Disabled0.020.040.060.080.10.0890.087
OpenBenchmarking.orgMegahertz, More Is BetterHigh Performance Conjugate Gradient 3.1CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3391.74 / Max: 3749Min: 2550 / Avg: 3373.06 / Max: 3862
OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.89 / Avg: 270.47 / Max: 300.51Min: 32.73 / Avg: 275.44 / Max: 308.991. 3DV Disabled: Approximate power consumption of 207634 Joules per run.2. Default: Approximate power consumption of 208018 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Temperature MonitorDefault3DV Disabled1326395265Min: 35 / Avg: 50.4 / Max: 66.38Min: 36.38 / Avg: 51.21 / Max: 67.25
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 144 144 144 - RT: 60Default3DV Disabled612182430Min: 23.26 / Avg: 24.61 / Max: 27.42Min: 23.27 / Avg: 23.52 / Max: 23.951. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/scivis/real_timeDefault3DV Disabled612182430SE +/- 0.01, N = 3SE +/- 0.03, N = 325.1124.06
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.12Benchmark: particle_volume/scivis/real_time3DV DisabledDefault0.02360.04720.07080.09440.1180.1050.101
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay 2.12CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3311.04 / Max: 3694Min: 2550 / Avg: 3209.01 / Max: 3694
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.12CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 28.9 / Avg: 228.1 / Max: 274.6Min: 33.37 / Avg: 248.7 / Max: 303.351. 3DV Disabled: Approximate power consumption of 32770 Joules per run.2. Default: Approximate power consumption of 35564 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay 2.12CPU Temperature Monitor3DV DisabledDefault1326395265Min: 38 / Avg: 51.15 / Max: 58Min: 39.63 / Avg: 55.63 / Max: 64.5
OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/scivis/real_timeDefault3DV Disabled612182430Min: 25.1 / Avg: 25.11 / Max: 25.13Min: 24 / Avg: 24.06 / Max: 24.09

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DDefault3DV Disabled2K4K6K8K10KSE +/- 40.63, N = 4SE +/- 149.17, N = 1510697.9010257.811. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DDefault3DV Disabled112233445547.1444.15
OpenBenchmarking.orgMegahertz, More Is BetterNAS Parallel Benchmarks 3.4CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 2995.75 / Max: 3719Min: 2550 / Avg: 2948.69 / Max: 3701
OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 32.16 / Avg: 226.94 / Max: 305.64Min: 28.63 / Avg: 232.35 / Max: 316.871. Default: Approximate power consumption of 3971 Joules per run.2. 3DV Disabled: Approximate power consumption of 4182 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Temperature MonitorDefault3DV Disabled1224364860Min: 32.38 / Avg: 48.21 / Max: 58.5Min: 33.63 / Avg: 52.59 / Max: 62.75
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DDefault3DV Disabled2K4K6K8K10KMin: 10578.1 / Avg: 10697.9 / Max: 10753.6Min: 8759.6 / Avg: 10257.81 / Max: 10810.71. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/ao/real_timeDefault3DV Disabled612182430SE +/- 0.01, N = 3SE +/- 0.01, N = 325.1624.13
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.12Benchmark: particle_volume/ao/real_time3DV DisabledDefault0.02070.04140.06210.08280.10350.0920.087
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay 2.12CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3140.98 / Max: 3699Min: 2550 / Avg: 3043.1 / Max: 3696
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.12CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 28.41 / Avg: 260.95 / Max: 276.2Min: 31.96 / Avg: 288.36 / Max: 304.271. 3DV Disabled: Approximate power consumption of 25051 Joules per run.2. Default: Approximate power consumption of 27490 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay 2.12CPU Temperature Monitor3DV DisabledDefault1326395265Min: 32.25 / Avg: 54.32 / Max: 59.88Min: 33.5 / Avg: 58.22 / Max: 65.5
OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/ao/real_timeDefault3DV Disabled612182430Min: 25.15 / Avg: 25.16 / Max: 25.17Min: 24.11 / Avg: 24.13 / Max: 24.15

Timed LLVM Compilation

This test times how long it takes to compile/build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Unix MakefilesDefault3DV Disabled4080120160200SE +/- 1.07, N = 3SE +/- 1.10, N = 3184.05191.73
OpenBenchmarking.orgMegahertz, More Is BetterTimed LLVM Compilation 16.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 3481.06 / Max: 3718Min: 2550 / Avg: 3478.68 / Max: 3830
OpenBenchmarking.orgWatts, Fewer Is BetterTimed LLVM Compilation 16.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.11 / Avg: 199.12 / Max: 315.58Min: 32.62 / Avg: 205.06 / Max: 328.351. 3DV Disabled: Approximate power consumption of 41152 Joules per run.2. Default: Approximate power consumption of 40807 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterTimed LLVM Compilation 16.0CPU Temperature Monitor3DV DisabledDefault1326395265Min: 34.13 / Avg: 51.72 / Max: 66.88Min: 36.38 / Avg: 54.09 / Max: 68.25
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Unix MakefilesDefault3DV Disabled4080120160200Min: 182.2 / Avg: 184.05 / Max: 185.92Min: 189.54 / Avg: 191.73 / Max: 193.04

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUDefault3DV Disabled612182430SE +/- 0.10, N = 3SE +/- 0.19, N = 326.8925.861. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUDefault3DV Disabled612182430Min: 26.69 / Avg: 26.89 / Max: 27.04Min: 25.61 / Avg: 25.86 / Max: 26.241. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUDefault3DV Disabled400800120016002000SE +/- 6.69, N = 3SE +/- 13.28, N = 31766.541835.23MIN: 966.02 / MAX: 2507.19MIN: 923.09 / MAX: 2505.651. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgMegahertz, More Is BetterOpenVINO 2022.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 2900.85 / Max: 3701Min: 2550 / Avg: 2842 / Max: 3699
OpenBenchmarking.orgWatts, Fewer Is BetterOpenVINO 2022.3CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.89 / Avg: 307.94 / Max: 346.57Min: 33.65 / Avg: 315.19 / Max: 3521. 3DV Disabled: Approximate power consumption of 21350 Joules per run.2. Default: Approximate power consumption of 21958 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOpenVINO 2022.3CPU Temperature Monitor3DV DisabledDefault1428425670Min: 39.63 / Avg: 58.69 / Max: 67.25Min: 42 / Avg: 61.5 / Max: 74.88
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUDefault3DV Disabled30060090012001500Min: 1754.88 / Avg: 1766.54 / Max: 1778.04Min: 1808.7 / Avg: 1835.23 / Max: 1849.681. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-StreamDefault3DV Disabled306090120150SE +/- 0.07, N = 3SE +/- 0.06, N = 3114.18109.95
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-StreamDefault3DV Disabled20406080100Min: 114.09 / Avg: 114.18 / Max: 114.32Min: 109.83 / Avg: 109.95 / Max: 110.04

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Classroom - Compute: CPU-OnlyDefault3DV Disabled1020304050SE +/- 0.11, N = 3SE +/- 0.06, N = 340.5142.05
OpenBenchmarking.orgMegahertz, More Is BetterBlender 3.6CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2761.77 / Max: 3694Min: 2550 / Avg: 2644.99 / Max: 3701
OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.6CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.19 / Avg: 281.35 / Max: 312.83Min: 33.61 / Avg: 295.71 / Max: 327.321. 3DV Disabled: Approximate power consumption of 13036 Joules per run.2. Default: Approximate power consumption of 13208 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterBlender 3.6CPU Temperature Monitor3DV DisabledDefault1224364860Min: 37.75 / Avg: 56.6 / Max: 60.63Min: 39.13 / Avg: 59.32 / Max: 63.88
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Classroom - Compute: CPU-OnlyDefault3DV Disabled918273645Min: 40.36 / Avg: 40.51 / Max: 40.72Min: 41.97 / Avg: 42.05 / Max: 42.17

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.5Test: PUSCH Processor Benchmark, Throughput TotalDefault3DV Disabled4K8K12K16K20KSE +/- 23.08, N = 3SE +/- 16.77, N = 318408.117748.81. (CXX) g++ options: -march=native -mfma -O3 -fno-trapping-math -fno-math-errno -lgtest
OpenBenchmarking.orgMbps Per Watt, More Is BettersrsRAN Project 23.5Test: PUSCH Processor Benchmark, Throughput TotalDefault3DV Disabled142842567063.1262.08
OpenBenchmarking.orgMegahertz, More Is BettersrsRAN Project 23.5CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2532 / Avg: 2667.44 / Max: 3702Min: 2550 / Avg: 2626.77 / Max: 3703
OpenBenchmarking.orgWatts, Fewer Is BettersrsRAN Project 23.5CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 27.93 / Avg: 285.9 / Max: 322.14Min: 31.97 / Avg: 291.63 / Max: 328.781. 3DV Disabled: Approximate power consumption of 10483 Joules per run.2. Default: Approximate power consumption of 10499 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BettersrsRAN Project 23.5CPU Temperature Monitor3DV DisabledDefault1224364860Min: 28.13 / Avg: 52.01 / Max: 59.63Min: 28.5 / Avg: 53.38 / Max: 61.25
OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.5Test: PUSCH Processor Benchmark, Throughput TotalDefault3DV Disabled3K6K9K12K15KMin: 18369 / Avg: 18408.13 / Max: 18448.9Min: 17731.4 / Avg: 17748.77 / Max: 17782.31. (CXX) g++ options: -march=native -mfma -O3 -fno-trapping-math -fno-math-errno -lgtest

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-StreamDefault3DV Disabled90180270360450SE +/- 0.35, N = 3SE +/- 0.14, N = 3417.86433.14
OpenBenchmarking.orgMegahertz, More Is BetterNeural Magic DeepSparse 1.5CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3175.39 / Max: 3715Min: 2550 / Avg: 3125.48 / Max: 3701
OpenBenchmarking.orgWatts, Fewer Is BetterNeural Magic DeepSparse 1.5CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 30.46 / Avg: 238.28 / Max: 301.8Min: 33.19 / Avg: 242.34 / Max: 308.021. 3DV Disabled: Approximate power consumption of 13105 Joules per run.2. Default: Approximate power consumption of 13409 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNeural Magic DeepSparse 1.5CPU Temperature Monitor3DV DisabledDefault1326395265Min: 36.38 / Avg: 52.33 / Max: 64.25Min: 38.25 / Avg: 53.02 / Max: 62.5
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-StreamDefault3DV Disabled80160240320400Min: 417.31 / Avg: 417.86 / Max: 418.5Min: 432.87 / Avg: 433.14 / Max: 433.36

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: MediumDefault3DV Disabled90180270360450SE +/- 0.05, N = 8SE +/- 0.30, N = 8419.43404.661. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.0Preset: Medium3DV DisabledDefault0.9021.8042.7063.6084.514.0093.896
OpenBenchmarking.orgMegahertz, More Is BetterASTC Encoder 4.0CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2446 / Avg: 3096.87 / Max: 3698Min: 2550 / Avg: 3072.03 / Max: 3700
OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 27.88 / Avg: 100.95 / Max: 305.62Min: 31.68 / Avg: 107.65 / Max: 325.071. 3DV Disabled: Approximate power consumption of 858 Joules per run.2. Default: Approximate power consumption of 915 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterASTC Encoder 4.0CPU Temperature Monitor3DV DisabledDefault1020304050Min: 28.38 / Avg: 34.14 / Max: 48.38Min: 29 / Avg: 34.66 / Max: 48.38
OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: MediumDefault3DV Disabled70140210280350Min: 419.17 / Avg: 419.43 / Max: 419.6Min: 403.61 / Avg: 404.66 / Max: 406.51. (CXX) g++ options: -O3 -flto -pthread

Numpy Benchmark

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

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkDefault3DV Disabled130260390520650SE +/- 0.96, N = 3SE +/- 2.45, N = 3586.50565.91
OpenBenchmarking.orgScore Per Watt, More Is BetterNumpy Benchmark3DV DisabledDefault1.23622.47243.70864.94486.1815.4945.478
OpenBenchmarking.orgMegahertz, More Is BetterNumpy BenchmarkCPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3671.41 / Max: 3733Min: 2550 / Avg: 3669.58 / Max: 3738
OpenBenchmarking.orgWatts, Fewer Is BetterNumpy BenchmarkCPU Power Consumption Monitor3DV DisabledDefault20406080100Min: 28.13 / Avg: 103 / Max: 110.77Min: 32.12 / Avg: 107.07 / Max: 114.311. 3DV Disabled: Approximate power consumption of 16378 Joules per run.2. Default: Approximate power consumption of 16918 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNumpy BenchmarkCPU Temperature Monitor3DV DisabledDefault1224364860Min: 29.5 / Avg: 35.58 / Max: 44.13Min: 30.88 / Avg: 42.07 / Max: 61.88
OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkDefault3DV Disabled100200300400500Min: 585.33 / Avg: 586.5 / Max: 588.41Min: 563.04 / Avg: 565.91 / Max: 570.79

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUDefault3DV Disabled246810SE +/- 0.00, N = 3SE +/- 0.00, N = 38.368.66MIN: 4.99 / MAX: 31.02MIN: 5.11 / MAX: 29.491. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgMegahertz, More Is BetterOpenVINO 2022.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 2761.96 / Max: 3699Min: 2550 / Avg: 2650.55 / Max: 3705
OpenBenchmarking.orgWatts, Fewer Is BetterOpenVINO 2022.3CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 33.48 / Avg: 303.32 / Max: 329.87Min: 29.76 / Avg: 303.52 / Max: 330.121. Default: Approximate power consumption of 19918 Joules per run.2. 3DV Disabled: Approximate power consumption of 19931 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOpenVINO 2022.3CPU Temperature Monitor3DV DisabledDefault1326395265Min: 41.75 / Avg: 60.11 / Max: 63.88Min: 42 / Avg: 60.22 / Max: 68.75
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUDefault3DV Disabled3691215Min: 8.36 / Avg: 8.36 / Max: 8.37Min: 8.65 / Avg: 8.66 / Max: 8.661. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeDefault3DV Disabled612182430SE +/- 0.02, N = 3SE +/- 0.01, N = 326.5125.60
OpenBenchmarking.orgItems Per Second Per Watt, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time3DV DisabledDefault0.0230.0460.0690.0920.1150.1020.099
OpenBenchmarking.orgMegahertz, More Is BetterOSPRay 2.12CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3123.87 / Max: 3701Min: 2550 / Avg: 3070.26 / Max: 3702
OpenBenchmarking.orgWatts, Fewer Is BetterOSPRay 2.12CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 29.05 / Avg: 249.93 / Max: 288.35Min: 33.33 / Avg: 266.48 / Max: 305.531. 3DV Disabled: Approximate power consumption of 13829 Joules per run.2. Default: Approximate power consumption of 14745 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOSPRay 2.12CPU Temperature Monitor3DV DisabledDefault1224364860Min: 38.75 / Avg: 54.05 / Max: 58.5Min: 40.13 / Avg: 56.83 / Max: 61.75
OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeDefault3DV Disabled612182430Min: 26.48 / Avg: 26.51 / Max: 26.54Min: 25.57 / Avg: 25.6 / Max: 25.62

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUDefault3DV Disabled12002400360048006000SE +/- 2.96, N = 3SE +/- 2.81, N = 35732.755538.581. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUDefault3DV Disabled10002000300040005000Min: 5726.94 / Avg: 5732.75 / Max: 5736.61Min: 5533.58 / Avg: 5538.58 / Max: 5543.321. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

Algebraic Multi-Grid Benchmark

AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2Default3DV Disabled500M1000M1500M2000M2500MSE +/- 4643811.51, N = 3SE +/- 1555438.45, N = 3241428300023327436671. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi
OpenBenchmarking.orgFigure Of Merit Per Watt, More Is BetterAlgebraic Multi-Grid Benchmark 1.23DV DisabledDefault2M4M6M8M10M10022771.119962692.30
OpenBenchmarking.orgMegahertz, More Is BetterAlgebraic Multi-Grid Benchmark 1.2CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3252.28 / Max: 3695Min: 2550 / Avg: 3226.98 / Max: 3738
OpenBenchmarking.orgWatts, Fewer Is BetterAlgebraic Multi-Grid Benchmark 1.2CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.82 / Avg: 232.74 / Max: 297.77Min: 32.75 / Avg: 242.33 / Max: 307.071. 3DV Disabled: Approximate power consumption of 6905 Joules per run.2. Default: Approximate power consumption of 7108 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterAlgebraic Multi-Grid Benchmark 1.2CPU Temperature Monitor3DV DisabledDefault1224364860Min: 36.63 / Avg: 48.3 / Max: 59.5Min: 38.75 / Avg: 49.63 / Max: 60.38
OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2Default3DV Disabled400M800M1200M1600M2000MMin: 2405449000 / Avg: 2414283000 / Max: 2421183000Min: 2329778000 / Avg: 2332743666.67 / Max: 23350400001. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: SemaphoresDefault3DV Disabled50M100M150M200M250MSE +/- 2019374.88, N = 3SE +/- 2598030.86, N = 3223213939.86215763708.611. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Semaphores3DV DisabledDefault200K400K600K800K1000K779203.94769507.51
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3052.85 / Max: 3694Min: 2550 / Avg: 3026.21 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 31.23 / Avg: 276.9 / Max: 315.09Min: 35.1 / Avg: 290.07 / Max: 328.091. 3DV Disabled: Approximate power consumption of 9415 Joules per run.2. Default: Approximate power consumption of 9766 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1326395265Min: 39.63 / Avg: 59.02 / Max: 63.5Min: 39.13 / Avg: 61.22 / Max: 68.13
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: SemaphoresDefault3DV Disabled40M80M120M160M200MMin: 220797987.26 / Avg: 223213939.86 / Max: 227224772.65Min: 210671957.14 / Avg: 215763708.61 / Max: 219206714.891. (CXX) g++ options: -O2 -std=gnu99 -lc

TensorFlow

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

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 512 - Model: GoogLeNetDefault3DV Disabled90180270360450SE +/- 3.82, N = 12SE +/- 5.03, N = 12409.02395.85
OpenBenchmarking.orgimages/sec Per Watt, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 512 - Model: GoogLeNet3DV DisabledDefault0.33320.66640.99961.33281.6661.4811.467
OpenBenchmarking.orgMegahertz, More Is BetterTensorFlow 2.12CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2396 / Avg: 3199.05 / Max: 4260Min: 2276 / Avg: 3192.29 / Max: 3898
OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow 2.12CPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 29.32 / Avg: 267.33 / Max: 300.15Min: 32.91 / Avg: 278.88 / Max: 305.161. 3DV Disabled: Approximate power consumption of 38941 Joules per run.2. Default: Approximate power consumption of 39811 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterTensorFlow 2.12CPU Temperature Monitor3DV DisabledDefault1326395265Min: 36.38 / Avg: 52.64 / Max: 61.25Min: 37.75 / Avg: 54.85 / Max: 66.25
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 512 - Model: GoogLeNetDefault3DV Disabled70140210280350Min: 401.46 / Avg: 409.02 / Max: 448.44Min: 387.95 / Avg: 395.85 / Max: 450.65

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Pabellon Barcelona - Compute: CPU-OnlyDefault3DV Disabled1224364860SE +/- 0.08, N = 3SE +/- 0.30, N = 349.6151.23
OpenBenchmarking.orgMegahertz, More Is BetterBlender 3.6CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 2736.53 / Max: 3695Min: 2550 / Avg: 2676.63 / Max: 3789
OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.6CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.75 / Avg: 291.31 / Max: 324.33Min: 33.82 / Avg: 309.97 / Max: 342.691. 3DV Disabled: Approximate power consumption of 16119 Joules per run.2. Default: Approximate power consumption of 16635 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterBlender 3.6CPU Temperature Monitor3DV DisabledDefault1326395265Min: 41.5 / Avg: 59.46 / Max: 69.38Min: 43.38 / Avg: 62.39 / Max: 65.75
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Pabellon Barcelona - Compute: CPU-OnlyDefault3DV Disabled1020304050Min: 49.52 / Avg: 49.61 / Max: 49.77Min: 50.89 / Avg: 51.23 / Max: 51.82

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Fishy Cat - Compute: CPU-OnlyDefault3DV Disabled510152025SE +/- 0.09, N = 3SE +/- 0.04, N = 320.6221.29
OpenBenchmarking.orgMegahertz, More Is BetterBlender 3.6CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2764.16 / Max: 3694Min: 2550 / Avg: 2693.14 / Max: 3705
OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.6CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.13 / Avg: 260.89 / Max: 323.88Min: 33.77 / Avg: 271.88 / Max: 341.281. 3DV Disabled: Approximate power consumption of 6696 Joules per run.2. Default: Approximate power consumption of 6797 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterBlender 3.6CPU Temperature Monitor3DV DisabledDefault1428425670Min: 39.38 / Avg: 54.39 / Max: 59.88Min: 41 / Avg: 56.75 / Max: 71.5
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Fishy Cat - Compute: CPU-OnlyDefault3DV Disabled510152025Min: 20.45 / Avg: 20.62 / Max: 20.73Min: 21.22 / Avg: 21.29 / Max: 21.33

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 TimeDefault3DV Disabled20406080100108.37111.441. (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

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 160 160 160 - RT: 60Default3DV Disabled612182430SE +/- 0.34, N = 3SE +/- 0.23, N = 923.8423.231. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 160 160 160 - RT: 60Default3DV Disabled0.01960.03920.05880.07840.0980.0870.086
OpenBenchmarking.orgMegahertz, More Is BetterHigh Performance Conjugate Gradient 3.1CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3383.7 / Max: 3736Min: 2550 / Avg: 3369.92 / Max: 3898
OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.75 / Avg: 270.81 / Max: 310.7Min: 17.36 / Avg: 275.05 / Max: 305.311. 3DV Disabled: Approximate power consumption of 289374 Joules per run.2. Default: Approximate power consumption of 289629 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Temperature MonitorDefault3DV Disabled1326395265Min: 35.38 / Avg: 50.61 / Max: 65Min: 36.88 / Avg: 51.12 / Max: 69.13
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 160 160 160 - RT: 60Default3DV Disabled612182430Min: 23.31 / Avg: 23.84 / Max: 24.48Min: 22.25 / Avg: 23.23 / Max: 23.891. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: CPU StressDefault3DV Disabled50K100K150K200K250KSE +/- 233.96, N = 3SE +/- 296.75, N = 3212380.76206993.591. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: CPU Stress3DV DisabledDefault160320480640800721.83719.16
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2945 / Avg: 3062.44 / Max: 3738Min: 2550 / Avg: 3044.93 / Max: 3698
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 103.46 / Avg: 286.76 / Max: 314.85Min: 34.32 / Avg: 295.32 / Max: 331.911. 3DV Disabled: Approximate power consumption of 9750 Joules per run.2. Default: Approximate power consumption of 10139 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1326395265Min: 38.5 / Avg: 57.49 / Max: 63.63Min: 38 / Avg: 57.53 / Max: 64.38
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: CPU StressDefault3DV Disabled40K80K120K160K200KMin: 212019.7 / Avg: 212380.76 / Max: 212819.06Min: 206450.45 / Avg: 206993.59 / Max: 207472.361. (CXX) g++ options: -O2 -std=gnu99 -lc

Timed PHP Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 8.1.9Time To CompileDefault3DV Disabled816243240SE +/- 0.27, N = 3SE +/- 0.29, N = 333.4534.31
OpenBenchmarking.orgMegahertz, More Is BetterTimed PHP Compilation 8.1.9CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3578.73 / Max: 3719Min: 2550 / Avg: 3570.29 / Max: 3709
OpenBenchmarking.orgWatts, Fewer Is BetterTimed PHP Compilation 8.1.9CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.81 / Avg: 122.72 / Max: 307.64Min: 32.2 / Avg: 128.15 / Max: 316.611. 3DV Disabled: Approximate power consumption of 5154 Joules per run.2. Default: Approximate power consumption of 5297 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterTimed PHP Compilation 8.1.9CPU Temperature MonitorDefault3DV Disabled1224364860Min: 33.63 / Avg: 42.92 / Max: 62.75Min: 32.25 / Avg: 43.28 / Max: 57.38
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 8.1.9Time To CompileDefault3DV Disabled714212835Min: 33.1 / Avg: 33.45 / Max: 33.99Min: 33.97 / Avg: 34.31 / Max: 34.88

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUDefault3DV Disabled612182430SE +/- 0.10, N = 3SE +/- 0.11, N = 327.1126.431. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUDefault3DV Disabled612182430Min: 26.94 / Avg: 27.11 / Max: 27.3Min: 26.29 / Avg: 26.43 / Max: 26.651. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: MutexDefault3DV Disabled11M22M33M44M55MSE +/- 152725.58, N = 3SE +/- 127023.50, N = 349736118.0648488923.411. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Mutex3DV DisabledDefault40K80K120K160K200K173729.31171399.96
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2958.29 / Max: 3701Min: 2550 / Avg: 2865.47 / Max: 3686
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.46 / Avg: 279.11 / Max: 317.38Min: 32.15 / Avg: 290.18 / Max: 326.951. 3DV Disabled: Approximate power consumption of 9490 Joules per run.2. Default: Approximate power consumption of 9866 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1224364860Min: 32.75 / Avg: 53.51 / Max: 61.25Min: 32.75 / Avg: 55.69 / Max: 62.25
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: MutexDefault3DV Disabled9M18M27M36M45MMin: 49504282.89 / Avg: 49736118.06 / Max: 50024269.94Min: 48329712.64 / Avg: 48488923.41 / Max: 48739975.161. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUDefault3DV Disabled400800120016002000SE +/- 4.64, N = 3SE +/- 6.35, N = 31753.631796.42MIN: 955.99 / MAX: 2502.09MIN: 961.13 / MAX: 25031. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgMegahertz, More Is BetterOpenVINO 2022.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 2875.18 / Max: 3702Min: 2550 / Avg: 2823.57 / Max: 3695
OpenBenchmarking.orgWatts, Fewer Is BetterOpenVINO 2022.3CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.9 / Avg: 309.37 / Max: 347.96Min: 33.85 / Avg: 313.74 / Max: 350.531. 3DV Disabled: Approximate power consumption of 21656 Joules per run.2. Default: Approximate power consumption of 21857 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOpenVINO 2022.3CPU Temperature Monitor3DV DisabledDefault1530456075Min: 41.5 / Avg: 60.11 / Max: 67.25Min: 42.25 / Avg: 61.64 / Max: 76.63
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUDefault3DV Disabled30060090012001500Min: 1745.5 / Avg: 1753.63 / Max: 1761.58Min: 1784.22 / Avg: 1796.42 / Max: 1805.561. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 192 - Buffer Length: 256 - Filter Length: 512Default3DV Disabled300M600M900M1200M1500MSE +/- 762306.44, N = 3SE +/- 1211518.79, N = 3128786666712573666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 1.6Threads: 192 - Buffer Length: 256 - Filter Length: 512Default3DV Disabled1000K2000K3000K4000K5000K4769331.184758956.49
OpenBenchmarking.orgMegahertz, More Is BetterLiquid-DSP 1.6CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2775.37 / Max: 3696Min: 2550 / Avg: 2709.52 / Max: 3694
OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 1.6CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.4 / Avg: 264.21 / Max: 299.88Min: 33.4 / Avg: 270.03 / Max: 308.851. 3DV Disabled: Approximate power consumption of 9071 Joules per run.2. Default: Approximate power consumption of 9271 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterLiquid-DSP 1.6CPU Temperature Monitor3DV DisabledDefault1224364860Min: 36.88 / Avg: 53.63 / Max: 58Min: 37.38 / Avg: 54.69 / Max: 59.5
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 192 - Buffer Length: 256 - Filter Length: 512Default3DV Disabled200M400M600M800M1000MMin: 1286700000 / Avg: 1287866666.67 / Max: 1289300000Min: 1255000000 / Avg: 1257366666.67 / Max: 12590000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Second, More Is BetterASKAP 1.0Test: Hogbom Clean OpenMPDefault3DV Disabled30060090012001500SE +/- 4.24, N = 4SE +/- 6.92, N = 41212.171183.561. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgIterations Per Second Per Watt, More Is BetterASKAP 1.0Test: Hogbom Clean OpenMPDefault3DV Disabled369121512.7112.51
OpenBenchmarking.orgMegahertz, More Is BetterASKAP 1.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3452.19 / Max: 3698Min: 2550 / Avg: 3434.5 / Max: 3704
OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0CPU Power Consumption Monitor3DV DisabledDefault4080120160200Min: 27.95 / Avg: 94.61 / Max: 215.24Min: 32.13 / Avg: 95.34 / Max: 220.011. 3DV Disabled: Approximate power consumption of 1703 Joules per run.2. Default: Approximate power consumption of 1621 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterASKAP 1.0CPU Temperature MonitorDefault3DV Disabled1020304050Min: 29.88 / Avg: 33.7 / Max: 38Min: 28.5 / Avg: 41.06 / Max: 46.88
OpenBenchmarking.orgIterations Per Second, More Is BetterASKAP 1.0Test: Hogbom Clean OpenMPDefault3DV Disabled2004006008001000Min: 1204.82 / Avg: 1212.17 / Max: 1219.51Min: 1162.79 / Avg: 1183.56 / Max: 1190.481. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector MathDefault3DV Disabled120K240K360K480K600KSE +/- 61.16, N = 3SE +/- 28.98, N = 3545725.73533171.171. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Vector Math3DV DisabledDefault4008001200160020002008.921970.94
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2802.38 / Max: 2899Min: 2550 / Avg: 2758.12 / Max: 3517
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 30.14 / Avg: 265.4 / Max: 301.19Min: 35.04 / Avg: 276.89 / Max: 313.941. 3DV Disabled: Approximate power consumption of 9024 Joules per run.2. Default: Approximate power consumption of 9414 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1224364860Min: 38.25 / Avg: 56.69 / Max: 61.13Min: 39.88 / Avg: 58.33 / Max: 62.25
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector MathDefault3DV Disabled90K180K270K360K450KMin: 545611.3 / Avg: 545725.73 / Max: 545820.37Min: 533114.27 / Avg: 533171.17 / Max: 533209.181. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Wide Vector MathDefault3DV Disabled700K1400K2100K2800K3500KSE +/- 964.27, N = 3SE +/- 2839.05, N = 33485374.333407854.771. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Wide Vector Math3DV DisabledDefault3K6K9K12K15K12778.5212416.66
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2660.68 / Max: 3732Min: 2550 / Avg: 2600.11 / Max: 3695
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.79 / Avg: 266.69 / Max: 311.55Min: 17.85 / Avg: 280.7 / Max: 326.391. 3DV Disabled: Approximate power consumption of 9423 Joules per run.2. Default: Approximate power consumption of 9731 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1224364860Min: 26.25 / Avg: 49.71 / Max: 58Min: 25.38 / Avg: 49.88 / Max: 60.88
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Wide Vector MathDefault3DV Disabled600K1200K1800K2400K3000KMin: 3483461.35 / Avg: 3485374.33 / Max: 3486542.55Min: 3402364.27 / Avg: 3407854.77 / Max: 3411853.61. (CXX) g++ options: -O2 -std=gnu99 -lc

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm.in input file (Problem 5). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsDefault3DV Disabled3691215SE +/- 0.04, N = 5SE +/- 0.03, N = 510.2910.481. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
OpenBenchmarking.orgMegahertz, More Is BetterCloverLeafCPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3223.89 / Max: 3696Min: 2550 / Avg: 3222.32 / Max: 3701
OpenBenchmarking.orgWatts, Fewer Is BetterCloverLeafCPU Power Consumption Monitor3DV DisabledDefault50100150200250Min: 28.76 / Avg: 191.22 / Max: 262.66Min: 32.22 / Avg: 198.33 / Max: 272.091. 3DV Disabled: Approximate power consumption of 2792 Joules per run.2. Default: Approximate power consumption of 2816 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterCloverLeafCPU Temperature Monitor3DV DisabledDefault1020304050Min: 33.13 / Avg: 42.29 / Max: 46.88Min: 34.13 / Avg: 43.52 / Max: 48.25
OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsDefault3DV Disabled3691215Min: 10.18 / Avg: 10.29 / Max: 10.39Min: 10.42 / Avg: 10.48 / Max: 10.561. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: BMW27 - Compute: CPU-OnlyDefault3DV Disabled48121620SE +/- 0.10, N = 3SE +/- 0.03, N = 316.2816.57
OpenBenchmarking.orgMegahertz, More Is BetterBlender 3.6CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 2762.29 / Max: 3694Min: 2550 / Avg: 2755 / Max: 3697
OpenBenchmarking.orgWatts, Fewer Is BetterBlender 3.6CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.51 / Avg: 248.46 / Max: 318.51Min: 33.71 / Avg: 262.38 / Max: 336.471. 3DV Disabled: Approximate power consumption of 5300 Joules per run.2. Default: Approximate power consumption of 5422 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterBlender 3.6CPU Temperature Monitor3DV DisabledDefault1428425670Min: 39.63 / Avg: 53.54 / Max: 59Min: 41 / Avg: 56.1 / Max: 71.5
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: BMW27 - Compute: CPU-OnlyDefault3DV Disabled48121620Min: 16.11 / Avg: 16.28 / Max: 16.46Min: 16.54 / Avg: 16.57 / Max: 16.62

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-StreamDefault3DV Disabled306090120150SE +/- 0.13, N = 3SE +/- 0.05, N = 3139.02141.31
OpenBenchmarking.orgMegahertz, More Is BetterNeural Magic DeepSparse 1.5CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3160.94 / Max: 3720Min: 2550 / Avg: 3135.29 / Max: 3695
OpenBenchmarking.orgWatts, Fewer Is BetterNeural Magic DeepSparse 1.5CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.79 / Avg: 264.48 / Max: 306.46Min: 33.12 / Avg: 266.28 / Max: 311.691. 3DV Disabled: Approximate power consumption of 11020 Joules per run.2. Default: Approximate power consumption of 11273 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNeural Magic DeepSparse 1.5CPU Temperature Monitor3DV DisabledDefault1326395265Min: 37.75 / Avg: 56.43 / Max: 63.63Min: 38.75 / Avg: 56.47 / Max: 64.88
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-StreamDefault3DV Disabled306090120150Min: 138.8 / Avg: 139.02 / Max: 139.24Min: 141.26 / Avg: 141.31 / Max: 141.4

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-StreamDefault3DV Disabled70140210280350SE +/- 0.41, N = 3SE +/- 0.08, N = 3344.41338.83
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-StreamDefault3DV Disabled60120180240300Min: 343.78 / Avg: 344.41 / Max: 345.18Min: 338.69 / Avg: 338.83 / Max: 338.97

Remhos

Remhos (REMap High-Order Solver) is a miniapp that solves the pure advection equations that are used to perform monotonic and conservative discontinuous field interpolation (remap) as part of the Eulerian phase in Arbitrary Lagrangian Eulerian (ALE) simulations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRemhos 1.0Test: Sample Remap ExampleDefault3DV Disabled3691215SE +/- 0.10, N = 6SE +/- 0.10, N = 710.2910.461. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgMegahertz, More Is BetterRemhos 1.0CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 2955.5 / Max: 3743Min: 2550 / Avg: 2924.66 / Max: 3701
OpenBenchmarking.orgWatts, Fewer Is BetterRemhos 1.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.73 / Avg: 227.05 / Max: 316.18Min: 33.49 / Avg: 228.9 / Max: 319.061. 3DV Disabled: Approximate power consumption of 3276 Joules per run.2. Default: Approximate power consumption of 3243 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterRemhos 1.0CPU Temperature Monitor3DV DisabledDefault1224364860Min: 36.38 / Avg: 52.63 / Max: 60.38Min: 41 / Avg: 55.58 / Max: 63.63
OpenBenchmarking.orgSeconds, Fewer Is BetterRemhos 1.0Test: Sample Remap ExampleDefault3DV Disabled3691215Min: 10.08 / Avg: 10.29 / Max: 10.72Min: 10.19 / Avg: 10.46 / Max: 10.971. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamDefault3DV Disabled2004006008001000SE +/- 0.46, N = 3SE +/- 0.28, N = 3797.95785.58
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamDefault3DV Disabled140280420560700Min: 797.12 / Avg: 797.95 / Max: 798.72Min: 785.2 / Avg: 785.58 / Max: 786.13

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamDefault3DV Disabled1428425670SE +/- 0.03, N = 3SE +/- 0.02, N = 360.0661.01
OpenBenchmarking.orgMegahertz, More Is BetterNeural Magic DeepSparse 1.5CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3168.66 / Max: 3716Min: 2550 / Avg: 3133.65 / Max: 3727
OpenBenchmarking.orgWatts, Fewer Is BetterNeural Magic DeepSparse 1.5CPU Power Consumption MonitorDefault3DV Disabled50100150200250Min: 32.86 / Avg: 251.27 / Max: 299.65Min: 30.24 / Avg: 251.88 / Max: 291.971. Default: Approximate power consumption of 10888 Joules per run.2. 3DV Disabled: Approximate power consumption of 10579 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNeural Magic DeepSparse 1.5CPU Temperature MonitorDefault3DV Disabled1326395265Min: 38.75 / Avg: 55.39 / Max: 63.88Min: 39.63 / Avg: 56.43 / Max: 65.38
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamDefault3DV Disabled1224364860Min: 60.03 / Avg: 60.06 / Max: 60.12Min: 60.97 / Avg: 61.01 / Max: 61.04

Timed Gem5 Compilation

This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 21.2Time To CompileDefault3DV Disabled306090120150SE +/- 0.72, N = 3SE +/- 1.60, N = 3137.69139.70
OpenBenchmarking.orgMegahertz, More Is BetterTimed Gem5 Compilation 21.2CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 3504.52 / Max: 3740Min: 2550 / Avg: 3493.88 / Max: 4264
OpenBenchmarking.orgWatts, Fewer Is BetterTimed Gem5 Compilation 21.2CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.01 / Avg: 154.4 / Max: 316.36Min: 34.02 / Avg: 160.58 / Max: 329.781. 3DV Disabled: Approximate power consumption of 22491 Joules per run.2. Default: Approximate power consumption of 22802 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterTimed Gem5 Compilation 21.2CPU Temperature Monitor3DV DisabledDefault1326395265Min: 31.25 / Avg: 44.33 / Max: 69.25Min: 33.13 / Avg: 45.47 / Max: 66.88
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 21.2Time To CompileDefault3DV Disabled306090120150Min: 136.56 / Avg: 137.69 / Max: 139.02Min: 136.77 / Avg: 139.7 / Max: 142.29

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUDefault3DV Disabled12002400360048006000SE +/- 0.38, N = 3SE +/- 2.58, N = 35672.065603.341. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUDefault3DV Disabled10002000300040005000Min: 5671.41 / Avg: 5672.06 / Max: 5672.74Min: 5598.18 / Avg: 5603.34 / Max: 5605.941. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUDefault3DV Disabled246810SE +/- 0.00, N = 3SE +/- 0.00, N = 38.458.55MIN: 4.5 / MAX: 30.54MIN: 4.32 / MAX: 31.021. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF
OpenBenchmarking.orgMegahertz, More Is BetterOpenVINO 2022.3CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2491 / Avg: 2550.52 / Max: 3695Min: 2489 / Avg: 2517.73 / Max: 3695
OpenBenchmarking.orgWatts, Fewer Is BetterOpenVINO 2022.3CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.29 / Avg: 292.24 / Max: 315.39Min: 33.77 / Avg: 298.63 / Max: 321.021. 3DV Disabled: Approximate power consumption of 18898 Joules per run.2. Default: Approximate power consumption of 19212 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterOpenVINO 2022.3CPU Temperature Monitor3DV DisabledDefault1224364860Min: 41 / Avg: 57.97 / Max: 61.63Min: 41.13 / Avg: 58.77 / Max: 62.25
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUDefault3DV Disabled3691215Min: 8.45 / Avg: 8.45 / Max: 8.45Min: 8.55 / Avg: 8.55 / Max: 8.561. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 192 192 192 - RT: 60Default3DV Disabled510152025SE +/- 0.18, N = 3SE +/- 0.20, N = 922.8322.571. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 192 192 192 - RT: 603DV DisabledDefault0.01870.03740.05610.07480.09350.0830.083
OpenBenchmarking.orgMegahertz, More Is BetterHigh Performance Conjugate Gradient 3.1CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3372.11 / Max: 3761Min: 2550 / Avg: 3354.37 / Max: 3701
OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.68 / Avg: 270.87 / Max: 318.01Min: 32.9 / Avg: 274.93 / Max: 314.31. 3DV Disabled: Approximate power consumption of 512447 Joules per run.2. Default: Approximate power consumption of 520815 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Temperature MonitorDefault3DV Disabled1326395265Min: 35.13 / Avg: 50.85 / Max: 67.25Min: 36.38 / Avg: 51.12 / Max: 68.63
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 192 192 192 - RT: 60Default3DV Disabled510152025Min: 22.49 / Avg: 22.83 / Max: 23.05Min: 22.18 / Avg: 22.57 / Max: 23.641. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

Ngspice

Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670Default3DV Disabled306090120150SE +/- 0.26, N = 3SE +/- 0.47, N = 3118.39119.711. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE
OpenBenchmarking.orgMegahertz, More Is BetterNgspice 34CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2550 / Avg: 3665.07 / Max: 3738Min: 2550 / Avg: 3660.52 / Max: 3725
OpenBenchmarking.orgWatts, Fewer Is BetterNgspice 34CPU Power Consumption Monitor3DV DisabledDefault20406080100Min: 27.75 / Avg: 100.59 / Max: 103.65Min: 31.73 / Avg: 104.98 / Max: 108.061. 3DV Disabled: Approximate power consumption of 12473 Joules per run.2. Default: Approximate power consumption of 12912 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterNgspice 34CPU Temperature Monitor3DV DisabledDefault816243240Min: 27.63 / Avg: 33.83 / Max: 40.38Min: 29 / Avg: 35.02 / Max: 40.38
OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670Default3DV Disabled20406080100Min: 117.87 / Avg: 118.39 / Max: 118.67Min: 118.76 / Avg: 119.7 / Max: 120.21. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

CPU Temperature Monitor

OpenBenchmarking.orgCelsiusCPU Temperature MonitorPhoronix Test Suite System Monitoring3DV DisabledDefault1530456075Min: 24.5 / Avg: 52.16 / Max: 78.13Min: 23.88 / Avg: 54.01 / Max: 78.13

CPU Power Consumption Monitor

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System Monitoring3DV DisabledDefault90180270360450Min: 14.18 / Avg: 250.73 / Max: 502.24Min: 16.08 / Avg: 255.79 / Max: 446.59

CPU Peak Freq (Highest CPU Core Frequency) Monitor

OpenBenchmarking.orgMegahertzCPU Peak Freq (Highest CPU Core Frequency) MonitorPhoronix Test Suite System MonitoringDefault3DV Disabled7001400210028003500Min: 2272 / Avg: 3283.5 / Max: 4260Min: 2227 / Avg: 3284.82 / Max: 4264

PETSc

OpenBenchmarking.orgCelsius, Fewer Is BetterPETSc 3.19CPU Temperature Monitor3DV DisabledDefault1530456075Min: 39.13 / Avg: 54.36 / Max: 74.25Min: 40.13 / Avg: 57.4 / Max: 78.13

OpenBenchmarking.orgWatts, Fewer Is BetterPETSc 3.19CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.73 / Avg: 255.11 / Max: 315.74Min: 33.57 / Avg: 266.45 / Max: 323.57

OpenBenchmarking.orgMegahertz, More Is BetterPETSc 3.19CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 3509.68 / Max: 4112Min: 2550 / Avg: 3509.24 / Max: 3768

OpenBenchmarking.orgMB/s Per Watt, More Is BetterPETSc 3.19Test: Streams3DV DisabledDefault20040060080010001040.531023.14

OpenBenchmarking.orgMB/s, More Is BetterPETSc 3.19Test: StreamsDefault3DV Disabled60K120K180K240K300KSE +/- 6007.86, N = 9SE +/- 799.96, N = 3272616.80265446.591. (CC) gcc options: -fPIC -O3 -O2 -lpthread -ludev -lpciaccess -lm
OpenBenchmarking.orgMB/s Per Watt, More Is BetterPETSc 3.19Test: Streams3DV DisabledDefault20040060080010001040.531023.14
OpenBenchmarking.orgMegahertz, More Is BetterPETSc 3.19CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled7001400210028003500Min: 2550 / Avg: 3509.68 / Max: 4112Min: 2550 / Avg: 3509.24 / Max: 3768
OpenBenchmarking.orgWatts, Fewer Is BetterPETSc 3.19CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.73 / Avg: 255.11 / Max: 315.74Min: 33.57 / Avg: 266.45 / Max: 323.571. 3DV Disabled: Approximate power consumption of 284785 Joules per run.2. Default: Approximate power consumption of 271425 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterPETSc 3.19CPU Temperature Monitor3DV DisabledDefault1530456075Min: 39.13 / Avg: 54.36 / Max: 74.25Min: 40.13 / Avg: 57.4 / Max: 78.13
OpenBenchmarking.orgMB/s, More Is BetterPETSc 3.19Test: StreamsDefault3DV Disabled50K100K150K200K250KMin: 224901.08 / Avg: 272616.8 / Max: 281195.62Min: 264130.7 / Avg: 265446.59 / Max: 266892.651. (CC) gcc options: -fPIC -O3 -O2 -lpthread -ludev -lpciaccess -lm

Stress-NG

OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1122334455Min: 37.25 / Avg: 46.74 / Max: 56Min: 38.75 / Avg: 49.85 / Max: 55.25

OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.9 / Avg: 223.12 / Max: 295.6Min: 34.23 / Avg: 263.23 / Max: 309.1

OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3377.03 / Max: 3698Min: 2550 / Avg: 3242.22 / Max: 3695

OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Matrix 3D MathDefault3DV Disabled142842567063.0536.56

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Matrix 3D MathDefault3DV Disabled4K8K12K16K20KSE +/- 154.68, N = 3SE +/- 866.14, N = 1216595.288156.461. (CXX) g++ options: -O2 -std=gnu99 -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.15.10Test: Matrix 3D MathDefault3DV Disabled142842567063.0536.56
OpenBenchmarking.orgMegahertz, More Is BetterStress-NG 0.15.10CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault6001200180024003000Min: 2550 / Avg: 3377.03 / Max: 3698Min: 2550 / Avg: 3242.22 / Max: 3695
OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.15.10CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.9 / Avg: 223.12 / Max: 295.6Min: 34.23 / Avg: 263.23 / Max: 309.11. 3DV Disabled: Approximate power consumption of 6880 Joules per run.2. Default: Approximate power consumption of 8774 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStress-NG 0.15.10CPU Temperature Monitor3DV DisabledDefault1122334455Min: 37.25 / Avg: 46.74 / Max: 56Min: 38.75 / Avg: 49.85 / Max: 55.25
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Matrix 3D MathDefault3DV Disabled3K6K9K12K15KMin: 16286.44 / Avg: 16595.28 / Max: 16765.13Min: 6300.88 / Avg: 8156.46 / Max: 13519.861. (CXX) g++ options: -O2 -std=gnu99 -lc

ASKAP

OpenBenchmarking.orgCelsius, Fewer Is BetterASKAP 1.0CPU Temperature Monitor3DV DisabledDefault918273645Min: 29.88 / Avg: 34.72 / Max: 40.88Min: 31.25 / Avg: 36.6 / Max: 45.13

OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0CPU Power Consumption MonitorDefault3DV Disabled20406080100Min: 32.13 / Avg: 83.45 / Max: 130.34Min: 28.03 / Avg: 84.46 / Max: 128.08

OpenBenchmarking.orgMegahertz, More Is BetterASKAP 1.0CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3400.92 / Max: 3760Min: 2349 / Avg: 3291.36 / Max: 3704

OpenBenchmarking.orgMillion Grid Points Per Second Per Watt, More Is BetterASKAP 1.0Test: tConvolve OpenMP - DegriddingDefault3DV Disabled140280420560700660.90311.68

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - DegriddingDefault3DV Disabled12K24K36K48K60KSE +/- 1901.83, N = 7SE +/- 302.56, N = 855153.026323.01. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMillion Grid Points Per Second Per Watt, More Is BetterASKAP 1.0Test: tConvolve OpenMP - DegriddingDefault3DV Disabled140280420560700660.90311.68
OpenBenchmarking.orgMegahertz, More Is BetterASKAP 1.0CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3400.92 / Max: 3760Min: 2349 / Avg: 3291.36 / Max: 3704
OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0CPU Power Consumption MonitorDefault3DV Disabled20406080100Min: 32.13 / Avg: 83.45 / Max: 130.34Min: 28.03 / Avg: 84.46 / Max: 128.081. Default: Approximate power consumption of 858 Joules per run.2. 3DV Disabled: Approximate power consumption of 1605 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterASKAP 1.0CPU Temperature Monitor3DV DisabledDefault918273645Min: 29.88 / Avg: 34.72 / Max: 40.88Min: 31.25 / Avg: 36.6 / Max: 45.13
OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - DegriddingDefault3DV Disabled10K20K30K40K50KMin: 53251.2 / Avg: 55153.03 / Max: 66564Min: 24205.1 / Avg: 26323.04 / Max: 26625.61. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - GriddingDefault3DV Disabled6K12K18K24K30KSE +/- 0.00, N = 7SE +/- 460.19, N = 1226625.623436.71. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - GriddingDefault3DV Disabled5K10K15K20K25KMin: 26625.6 / Avg: 26625.6 / Max: 26625.6Min: 19018.3 / Avg: 23436.68 / Max: 24205.11. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgCelsius, Fewer Is BetterASKAP 1.0CPU Temperature Monitor3DV DisabledDefault1224364860Min: 35.38 / Avg: 47.68 / Max: 63Min: 35 / Avg: 48.25 / Max: 60.63

OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.96 / Avg: 230.62 / Max: 311.9Min: 32.52 / Avg: 250.03 / Max: 329.63

OpenBenchmarking.orgMegahertz, More Is BetterASKAP 1.0CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3380.68 / Max: 3773Min: 2550 / Avg: 3248.05 / Max: 3789

OpenBenchmarking.orgMpix/sec Per Watt, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingDefault3DV Disabled60120180240300292.87180.25

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingDefault3DV Disabled16K32K48K64K80KSE +/- 0.00, N = 3SE +/- 1120.94, N = 673226.741569.41. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMpix/sec Per Watt, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingDefault3DV Disabled60120180240300292.87180.25
OpenBenchmarking.orgMegahertz, More Is BetterASKAP 1.0CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3380.68 / Max: 3773Min: 2550 / Avg: 3248.05 / Max: 3789
OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 28.96 / Avg: 230.62 / Max: 311.9Min: 32.52 / Avg: 250.03 / Max: 329.631. 3DV Disabled: Approximate power consumption of 12069 Joules per run.2. Default: Approximate power consumption of 8418 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterASKAP 1.0CPU Temperature Monitor3DV DisabledDefault1224364860Min: 35.38 / Avg: 47.68 / Max: 63Min: 35 / Avg: 48.25 / Max: 60.63
OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingDefault3DV Disabled13K26K39K52K65KMin: 73226.7 / Avg: 73226.7 / Max: 73226.7Min: 39359.3 / Avg: 41569.4 / Max: 46996.21. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Stockfish

OpenBenchmarking.orgCelsius, Fewer Is BetterStockfish 15CPU Temperature MonitorDefault3DV Disabled1326395265Min: 40.13 / Avg: 62.95 / Max: 67.75Min: 38.25 / Avg: 63.28 / Max: 67.38

OpenBenchmarking.orgWatts, Fewer Is BetterStockfish 15CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.51 / Avg: 311.32 / Max: 339Min: 34.09 / Avg: 314.42 / Max: 349.07

OpenBenchmarking.orgMegahertz, More Is BetterStockfish 15CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2410 / Avg: 2782.31 / Max: 3697Min: 2371 / Avg: 2775.79 / Max: 3730

OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterStockfish 15Total TimeDefault3DV Disabled200K400K600K800K1000K945505.98917179.61

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 15Total TimeDefault3DV Disabled60M120M180M240M300MSE +/- 1047576.89, N = 3SE +/- 5254044.14, N = 122972898922855330401. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto=jobserver
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterStockfish 15Total TimeDefault3DV Disabled200K400K600K800K1000K945505.98917179.61
OpenBenchmarking.orgMegahertz, More Is BetterStockfish 15CPU Peak Freq (Highest CPU Core Frequency) MonitorDefault3DV Disabled6001200180024003000Min: 2410 / Avg: 2782.31 / Max: 3697Min: 2371 / Avg: 2775.79 / Max: 3730
OpenBenchmarking.orgWatts, Fewer Is BetterStockfish 15CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 29.51 / Avg: 311.32 / Max: 339Min: 34.09 / Avg: 314.42 / Max: 349.071. 3DV Disabled: Approximate power consumption of 22752 Joules per run.2. Default: Approximate power consumption of 24630 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterStockfish 15CPU Temperature MonitorDefault3DV Disabled1326395265Min: 40.13 / Avg: 62.95 / Max: 67.75Min: 38.25 / Avg: 63.28 / Max: 67.38
OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 15Total TimeDefault3DV Disabled50M100M150M200M250MMin: 295206762 / Avg: 297289892 / Max: 298525568Min: 241019358 / Avg: 285533040.33 / Max: 3128223601. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto=jobserver

Xcompact3d Incompact3d

OpenBenchmarking.orgCelsius, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Temperature Monitor3DV DisabledDefault1020304050Min: 32.88 / Avg: 39.22 / Max: 49.25Min: 34.5 / Avg: 40.38 / Max: 50.25

OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 32.43 / Avg: 133.37 / Max: 311.38Min: 28.33 / Avg: 134.51 / Max: 303.63

OpenBenchmarking.orgMegahertz, More Is BetterXcompact3d Incompact3d 2021-03-11CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3075.21 / Max: 3769Min: 2550 / Avg: 2944.13 / Max: 3789

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per DirectionDefault3DV Disabled0.55441.10881.66322.21762.772SE +/- 0.04209465, N = 15SE +/- 0.01883751, N = 152.175216132.463954731. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
OpenBenchmarking.orgMegahertz, More Is BetterXcompact3d Incompact3d 2021-03-11CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3075.21 / Max: 3769Min: 2550 / Avg: 2944.13 / Max: 3789
OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption MonitorDefault3DV Disabled60120180240300Min: 32.43 / Avg: 133.37 / Max: 311.38Min: 28.33 / Avg: 134.51 / Max: 303.631. Default: Approximate power consumption of 916 Joules per run.2. 3DV Disabled: Approximate power consumption of 968 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Temperature Monitor3DV DisabledDefault1020304050Min: 32.88 / Avg: 39.22 / Max: 49.25Min: 34.5 / Avg: 40.38 / Max: 50.25
OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per DirectionDefault3DV Disabled246810Min: 1.88 / Avg: 2.18 / Max: 2.39Min: 2.37 / Avg: 2.46 / Max: 2.611. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

High Performance Conjugate Gradient

OpenBenchmarking.orgCelsius, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Temperature MonitorDefault3DV Disabled1326395265Min: 23.88 / Avg: 50.27 / Max: 65Min: 25.75 / Avg: 51.33 / Max: 65.38

OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 20.76 / Avg: 270.95 / Max: 299.01Min: 31.2 / Avg: 275.23 / Max: 314.17

OpenBenchmarking.orgMegahertz, More Is BetterHigh Performance Conjugate Gradient 3.1CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3399.69 / Max: 3759Min: 2550 / Avg: 3376.94 / Max: 3764

OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 104 104 104 - RT: 60Default3DV Disabled0.02180.04360.06540.08720.1090.0970.089

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 104 104 104 - RT: 60Default3DV Disabled612182430SE +/- 0.82, N = 9SE +/- 0.34, N = 926.8324.201. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 104 104 104 - RT: 60Default3DV Disabled0.02180.04360.06540.08720.1090.0970.089
OpenBenchmarking.orgMegahertz, More Is BetterHigh Performance Conjugate Gradient 3.1CPU Peak Freq (Highest CPU Core Frequency) Monitor3DV DisabledDefault7001400210028003500Min: 2550 / Avg: 3399.69 / Max: 3759Min: 2550 / Avg: 3376.94 / Max: 3764
OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption Monitor3DV DisabledDefault60120180240300Min: 20.76 / Avg: 270.95 / Max: 299.01Min: 31.2 / Avg: 275.23 / Max: 314.171. 3DV Disabled: Approximate power consumption of 77220 Joules per run.2. Default: Approximate power consumption of 74679 Joules per run.
OpenBenchmarking.orgCelsius, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Temperature MonitorDefault3DV Disabled1326395265Min: 23.88 / Avg: 50.27 / Max: 65Min: 25.75 / Avg: 51.33 / Max: 65.38
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 104 104 104 - RT: 60Default3DV Disabled612182430Min: 23.61 / Avg: 26.82 / Max: 31.85Min: 23.09 / Avg: 24.2 / Max: 26.181. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

Geometric Mean Of All Test Results

OpenBenchmarking.orgGeometric Mean, More Is BetterGeometric Mean Of All Test ResultsResult Composite - AMD EPYC 9684X 3D V-Cache BenchmarkDefault3DV Disabled50100150200250206.10184.54

158 Results Shown

OpenFOAM
libxsmm
NAS Parallel Benchmarks
libxsmm
Stress-NG
OpenVINO:
  Vehicle Detection FP16 - CPU:
    ms
    FPS
ASKAP
HeFFTe - Highly Efficient FFT for Exascale
OpenFOAM
NAS Parallel Benchmarks:
  LU.C
  BT.C
LeelaChessZero
Xmrig
HeFFTe - Highly Efficient FFT for Exascale
NAS Parallel Benchmarks
libxsmm
Stress-NG
NAS Parallel Benchmarks:
  MG.C
  CG.C
HeFFTe - Highly Efficient FFT for Exascale:
  c2c - FFTW - double - 256
  r2c - FFTW - double - 512
PyHPC Benchmarks
LeelaChessZero
HeFFTe - Highly Efficient FFT for Exascale
Xcompact3d Incompact3d
Google Draco
Stress-NG
OpenFOAM
WRF
Stress-NG
GROMACS
Embree
Palabos
GPAW
Embree
libxsmm
Embree
LULESH
ASKAP
PyHPC Benchmarks
Stress-NG:
  Pipe
  Futex
ASKAP
Ngspice
ASTC Encoder
Stress-NG
Palabos
miniFE
Timed Linux Kernel Compilation
Neural Magic DeepSparse
ASTC Encoder
Neural Magic DeepSparse
ACES DGEMM
Timed Godot Game Engine Compilation
Timed LLVM Compilation
Palabos
OSPRay Studio
Stress-NG
OSPRay Studio:
  2 - 4K - 16 - Path Tracer
  1 - 4K - 16 - Path Tracer
  3 - 4K - 1 - Path Tracer
Google Draco
OSPRay Studio:
  3 - 4K - 32 - Path Tracer
  1 - 4K - 1 - Path Tracer
  2 - 4K - 32 - Path Tracer
Blender
OSPRay:
  gravity_spheres_volume/dim_512/ao/real_time
  gravity_spheres_volume/dim_512/scivis/real_time
OSPRay Studio
HeFFTe - Highly Efficient FFT for Exascale
OSPRay Studio
NAMD
Timed Node.js Compilation
Stress-NG
High Performance Conjugate Gradient
OSPRay
NAS Parallel Benchmarks
OSPRay
Timed LLVM Compilation
OpenVINO:
  Person Detection FP32 - CPU:
    FPS
    ms
Neural Magic DeepSparse
Blender
srsRAN Project
Neural Magic DeepSparse
ASTC Encoder
Numpy Benchmark
OpenVINO
OSPRay
OpenVINO
Algebraic Multi-Grid Benchmark
Stress-NG
TensorFlow
Blender:
  Pabellon Barcelona - CPU-Only
  Fishy Cat - CPU-Only
OpenFOAM
High Performance Conjugate Gradient
Stress-NG
Timed PHP Compilation
OpenVINO
Stress-NG
OpenVINO
Liquid-DSP
ASKAP
Stress-NG:
  Vector Math
  Wide Vector Math
CloverLeaf
Blender
Neural Magic DeepSparse:
  CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream:
    ms/batch
    items/sec
Remhos
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream:
    items/sec
    ms/batch
Timed Gem5 Compilation
OpenVINO:
  Vehicle Detection FP16-INT8 - CPU:
    FPS
    ms
High Performance Conjugate Gradient
Ngspice
CPU Temperature Monitor:
  Phoronix Test Suite System Monitoring:
    Celsius
    Watts
    Megahertz
  CPU Temp Monitor:
    Celsius
  CPU Power Consumption Monitor:
    Watts
  CPU Peak Freq (Highest CPU Core Frequency) Monitor:
    Megahertz
  Streams:
    MB/s Per Watt
PETSc
Stress-NG:
  CPU Temp Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  Matrix 3D Math
Stress-NG
ASKAP:
  CPU Temp Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  tConvolve OpenMP - Degridding
ASKAP:
  tConvolve OpenMP - Degridding
  tConvolve OpenMP - Gridding
ASKAP:
  CPU Temp Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  tConvolve MPI - Gridding
ASKAP
Stockfish:
  CPU Temp Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  Total Time
Stockfish
Xcompact3d Incompact3d:
  CPU Temp Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
Xcompact3d Incompact3d
High Performance Conjugate Gradient:
  CPU Temp Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  104 104 104 - 60
High Performance Conjugate Gradient
Geometric Mean Of All Test Results

Default

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101121
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 July 2023 20:53 by user phoronix.

3DV Disabled

Processor: AMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 22.04, Kernel: 5.19.0-41-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 1024x768

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa101121
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 19 July 2023 04:28 by user phoronix.