AMD EPYC 7F72 2P Linux 5.11 Perf Governor

2 x AMD EPYC 7F72 24-Core testing looking at CPU freq invariance on 5.11 with patch. CPU power consumption monitoring via AMD_Energy interface at 1 second polling. Additional data with CPUFreq performance governor included.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2101253-HA-AMDEPYC7F96
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:

AV1 3 Tests
Bioinformatics 3 Tests
BLAS (Basic Linear Algebra Sub-Routine) Tests 4 Tests
C++ Boost Tests 5 Tests
Chess Test Suite 4 Tests
Timed Code Compilation 5 Tests
C/C++ Compiler Tests 20 Tests
Compression Tests 2 Tests
CPU Massive 38 Tests
Creator Workloads 21 Tests
Cryptography 4 Tests
Database Test Suite 4 Tests
Encoding 5 Tests
Finance 2 Tests
Fortran Tests 7 Tests
Game Development 5 Tests
HPC - High Performance Computing 28 Tests
LAPACK (Linear Algebra Pack) Tests 2 Tests
Machine Learning 8 Tests
Molecular Dynamics 6 Tests
MPI Benchmarks 8 Tests
Multi-Core 35 Tests
NVIDIA GPU Compute 6 Tests
Intel oneAPI 4 Tests
OpenCL 2 Tests
OpenMPI Tests 15 Tests
Programmer / Developer System Benchmarks 9 Tests
Python 2 Tests
Quantum Mechanics 2 Tests
Raytracing 4 Tests
Renderers 9 Tests
Scientific Computing 15 Tests
Server 6 Tests
Server CPU Tests 22 Tests
Single-Threaded 6 Tests
Texture Compression 2 Tests
Video Encoding 5 Tests
Common Workstation Benchmarks 6 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

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
Perf-Per
Dollar
Date
Triggered
  Test
  Duration
Linux 5.11 Git
January 22
  15 Hours
Linux 5.11 Patched
January 23
  15 Hours, 14 Minutes
CPUFreq Performance
January 24
  16 Hours, 38 Minutes
Invert Hiding All Results Option
  15 Hours, 38 Minutes
Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):


AMD EPYC 7F72 2P Linux 5.11 Perf GovernorProcessorMotherboardChipsetMemoryDiskGraphicsNetworkMonitorOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen ResolutionLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance2 x AMD EPYC 7F72 24-Core @ 3.20GHz (48 Cores / 96 Threads)Supermicro H11DSi-NT v2.00 (2.1 BIOS)AMD Starship/Matisse16 x 8192 MB DDR4-3200MT/s HMA81GR7CJR8N-XN1000GB Western Digital WD_BLACK SN850 1TBASPEED2 x Intel 10G X550TUbuntu 20.105.11.0-051100rc4daily20210122-generic (x86_64) 20210121GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.9GCC 10.2.0ext41920x1080VE2285.11.0-rc4-max-boost-inv-patch (x86_64) 20210121OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Disk Details- NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Details- Linux 5.11 Git: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301034- Linux 5.11 Patched: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301034- CPUFreq Performance: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Details- OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.10)Python Details- Python 3.8.6Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Linux 5.11 GitLinux 5.11 PatchedCPUFreq PerformanceResult OverviewPhoronix Test Suite 10.2.2100%107%114%121%128%dav1dCpuminer-Optx265DaCapo BenchmarkTimed GDB GNU Debugger CompilationInfluxDBNebular Empirical Analysis ToolCLOMPFFTWLAMMPS Molecular Dynamics SimulatorQMCPACKZstd CompressiononeDNNQuantum ESPRESSOIORTTSIOD 3D RendererAI Benchmark Alpharav1eOSPrayRodiniaRedisSVT-VP9Himeno BenchmarkTimed Godot Game Engine CompilationFFTEYafaRayKeyDBJohn The RipperChaos Group V-RAYLeelaChessZeroTNNHigh Performance Conjugate GradientLULESHNAMDNAS Parallel BenchmarksASKAPTimed Linux Kernel CompilationStockfishBlogBenchOpenFOAMPOV-RayGPAWTachyonCython BenchmarkBlenderTimed LLVM CompilationBYTE Unix BenchmarkBuild2PrimesieveIntel Open Image DenoiseSVT-AV1SQLite SpeedtestONNX RuntimePlaidMLAlgebraic Multi-Grid BenchmarkTungsten RendererLuxCoreRenderTimed MrBayes AnalysissimdjsonASTC EncoderBRL-CADLZ4 CompressionGcrypt LibraryNumpy BenchmarkDolfynGROMACSEtcpakGoogle SynthMarkRELIONQuantLibGnuPGasmFishSwetTSCPHierarchical INTegrationFinanceBenchTensorFlow Lite

Linux 5.11 GitLinux 5.11 PatchedCPUFreq PerformancePer Watt Result OverviewPhoronix Test Suite 10.2.2100%105%110%116%121%ASKAPdav1dCpuminer-OptHigh Performance Conjugate GradientAI Benchmark AlphaKeyDBZstd CompressionFFTWLAMMPS Molecular Dynamics SimulatorCLOMPTTSIOD 3D RendererIORRedisOSPrayFFTESVT-VP9BlogBenchx265asmFishInfluxDBNAS Parallel BenchmarksHimeno BenchmarkLULESHChaos Group V-RAYQuantLibEtcpakJohn The RipperStockfishLeelaChessZeroBYTE Unix BenchmarkNumpy BenchmarkAlgebraic Multi-Grid BenchmarkLZ4 CompressionSwetONNX RuntimeGoogle SynthMarkHierarchical INTegrationBRL-CADTSCPPlaidMLSVT-AV1GROMACSLuxCoreRenderrav1eIntel Open Image DenoiseP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.M

AMD EPYC 7F72 2P Linux 5.11 Perf Governorqe: AUSURF112relion: Basic - CPUblogbench: Readlczero: BLASfftw: Float + SSE - 2D FFT Size 4096plaidml: No - Inference - VGG19 - CPUplaidml: No - Inference - ResNet 50 - CPUplaidml: No - Inference - VGG16 - CPUsvt-av1: Enc Mode 0 - 1080plczero: Eigenai-benchmark: Device AI Scoreai-benchmark: Device Inference Scoreai-benchmark: Device Training Scorekeydb: lammps: 20k Atomsbrl-cad: VGR Performance Metriconnx: super-resolution-10 - OpenMP CPUyafaray: Total Time For Sample Scenehint: FLOATtensorflow-lite: NASNet Mobileonednn: Recurrent Neural Network Training - f32 - CPUnumpy: onnx: yolov4 - OpenMP CPUgcrypt: build-llvm: Time To Compileopenfoam: Motorbike 60Mtensorflow-lite: Mobilenet Quantblender: Barbershop - CPU-Onlyhpcg: tensorflow-lite: SqueezeNetcpuminer-opt: Garlicoinasmfish: 1024 Hash Memory, 26 Depthv-ray: CPUtensorflow-lite: Mobilenet Floatcpuminer-opt: x25xcpuminer-opt: Skeincoinjohn-the-ripper: MD5influxdb: 4 - 10000 - 2,5000,1 - 10000byte: Dhrystone 2neat: cpuminer-opt: Quad SHA-256, Pyritetensorflow-lite: Inception ResNet V2rodinia: OpenMP HotSpot3Dttsiod-renderer: Phong Rendering With Soft-Shadow Mappingbuild-gdb: Time To Compiledav1d: Chimera 1080p 10-bitbuild-linux-kernel: Time To Compileinfluxdb: 64 - 10000 - 2,5000,1 - 10000redis: SETcompress-lz4: 3 - Decompression Speedcompress-lz4: 3 - Compression Speedior: 8MB - Default Test Directorymrbayes: Primate Phylogeny Analysisdacapobench: Tradebeanshimeno: Poisson Pressure Solverospray: San Miguel - Path Tracergnupg: 2.7GB Sample File Encryptionluxcorerender: Rainbow Colors and Prismqmcpack: simple-H2Osimdjson: Kostyasqlite-speedtest: Timed Time - Size 1,000compress-zstd: 3dacapobench: H2build2: Time To Compilefinancebench: Bonds OpenMPcompress-lz4: 9 - Compression Speedcompress-lz4: 9 - Decompression Speedluxcorerender: DLSCtensorflow-lite: Inception V4gpaw: Carbon Nanotubebuild-godot: Time To Compilesimdjson: LargeRandtungsten: Water Causticrav1e: 5rav1e: 1simdjson: PartialTweetssimdjson: DistinctUserIDrodinia: OpenMP Leukocyterodinia: OpenMP LavaMDclomp: Static OMP Speeduponednn: Deconvolution Batch shapes_1d - f32 - CPUfinancebench: Repo OpenMPredis: GETastcenc: Exhaustivesvt-av1: Enc Mode 8 - 1080prav1e: 6gromacs: Water Benchmarkospray: XFrog Forest - Path Tracerredis: SADDospray: San Miguel - SciVisnpb: LU.Cnpb: EP.Drodinia: OpenMP CFD Solverquantlib: stockfish: Total Timeswet: Averagex265: Bosphorus 1080petcpak: ETC2compress-lz4: 1 - Decompression Speedcompress-lz4: 1 - Compression Speedcpuminer-opt: LBC, LBRY Creditsx265: Bosphorus 4Kaskap: tConvolve MPI - Degriddingaskap: tConvolve MPI - Griddingrodinia: OpenMP Streamclusterjohn-the-ripper: Blowfishsynthmark: VoiceMark_100rav1e: 10openfoam: Motorbike 30Mcython-bench: N-Queensnamd: ATPase Simulation - 327,506 Atomsdav1d: Summer Nature 4Kospray: XFrog Forest - SciVisior: 2MB - Default Test Directorydacapobench: Jythonospray: NASA Streamlines - Path Tracertnn: CPU - MobileNet v2onednn: IP Shapes 1D - f32 - CPUsvt-av1: Enc Mode 4 - 1080petcpak: ETC1 + Ditheringamg: tachyon: Total Timetnn: CPU - SqueezeNet v1.1dolfyn: Computational Fluid Dynamicsredis: LPUSHetcpak: ETC1povray: Trace Timeastcenc: Thoroughlulesh: dacapobench: Tradesoaponednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUtungsten: Volumetric Causticffte: N=256, 3D Complex FFT Routinesvt-vp9: Visual Quality Optimized - Bosphorus 1080ponednn: IP Shapes 3D - f32 - CPUospray: Magnetic Reconnection - SciVisonednn: Convolution Batch Shapes Auto - f32 - CPUoidn: Memorialtungsten: Hairospray: NASA Streamlines - SciVisprimesieve: 1e12 Prime Number Generationsvt-vp9: VMAF Optimized - Bosphorus 1080pnpb: EP.Csvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080pospray: Magnetic Reconnection - Path Tracerlammps: Rhodopsin Proteintscp: AI Chess PerformanceLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance1217.49349.852108440541061846822.094.6525.120.0924284275616971059302893.5624.993638971439384.739322702844.182041897711317.40323.93175233.825210.347129.6745083.9158.6430.241565193.09937.311173109855480346659.21524.213637844550333807463.138181643.627.06729750276572697.942627.20897.641130.6125.9231231991.21380890.2210404.549.43531.0982.65059544233.5115834.3077.3028.7231.1770.5770.1608205.2531068.14157601.36328147.6310488.07.8189464060.70260.8580.3721.32531.0450.3680.620.6553.86252.79443.92.4054940124.3736981689203.1041.1768.2291.3705.2395.911539146.2152.63147443.863854.609.2552149.89794508968588886347.66155.18211334.09622.9913247718.6311870.37426.0811.20970824712.2522.90218.7126.9500.45451308.2911.11505.19489716.39303.4491.620887.663244.793143777133317.9375274.94418.7181220973.71266.31511.4855.6919576.12251700.5476745.30105174206.13000387311.850.88134832.260.91419828.096.6005471.434.540369.013788.88381.0825021.12911151331171.03348.294110311840611701522.494.6325.420.0914433278717201067294214.3725.077636521421087.143323144417.171971340441123.32323.00181232.542208.786128.2841034.0156.8330.826262195.49949.881176329555346039523.51541.763640174612308812193.638319339.824.63329699573628596.603655.22592.916133.3725.7521256112.11427348.1010666.048.95520.7282.04255914286.6283094.3277.1798.7629.2810.5770.5408270.5521767.32256769.45312547.7610489.87.8081075059.85159.1770.3621.33291.0680.3720.630.6552.68452.09247.82.3329039406.7578121711621.5240.9768.2311.4085.2615.951611164.3454.97154376.763863.458.8822157.29704260168748026249.45155.79811305.09757.1713903719.7411944.27453.5110.33872636714.9143.05418.3026.6040.44472317.4511.19475.25477816.39289.7641.554477.648245.595144871833318.0565274.86918.6521217218.75267.58711.3055.6519771.22351480.5219685.26235178738.12497094323.810.84924832.620.86378228.396.6903671.434.535364.813841.48371.4825023.78711145621249.06348.338108634741471733522.434.6625.220.0914450290817751133303171.3325.116635725419087.331322775283.801251328441247.48324.48185232.760208.043128.8141180.1158.0730.926061347.29964.351173708715501439981.91537.525229484762000956189.438656226.325.03742023173799396.885665.44885.277181.9525.4141360163.01454741.4210410.349.29539.6382.26146714135.5191544.3377.0228.7628.9710.5769.8837770.3457067.33456094.48828148.0210555.17.8981888759.75758.7460.3721.46521.0950.3750.630.6551.29052.36047.42.3046739948.3567711782755.9741.0768.4651.4465.2555.951610484.5055.56153770.573867.308.5102156.39608211468613210962.29155.78311150.69651.6119408720.7511492.97441.9510.43271757714.1813.17718.2926.8950.44469363.3311.19517.33474016.39297.1321.579637.488245.453144417466718.2004274.09718.6351251006.00267.13611.4275.6719334.86246210.5146045.31183180037.95358586309.610.81362832.800.86754528.336.6950571.434.489346.713857.33363.18333.3324.6391116255OpenBenchmarking.org

Quantum ESPRESSO

Quantum ESPRESSO is an integrated suite of Open-Source computer codes for electronic-structure calculations and materials modeling at the nanoscale. It is based on density-functional theory, plane waves, and pseudopotentials. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterQuantum ESPRESSO 6.7Input: AUSURF112Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance30060090012001500SE +/- 11.28, N = 3SE +/- 12.21, N = 4SE +/- 19.05, N = 91217.491171.031249.061. (F9X) gfortran options: -lopenblas -lFoX_dom -lFoX_sax -lFoX_wxml -lFoX_common -lFoX_utils -lFoX_fsys -lfftw3 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent -levent_pthreads -lutil -lm -lrt -lz
OpenBenchmarking.orgWatts, Fewer Is BetterQuantum ESPRESSO 6.7CPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance90180270360450Min: 73.52 / Avg: 387.82 / Max: 492.31Min: 120.56 / Avg: 385.69 / Max: 492.5Min: 70.27 / Avg: 389.94 / Max: 492.33
OpenBenchmarking.orgSeconds, Fewer Is BetterQuantum ESPRESSO 6.7Input: AUSURF112Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance2004006008001000Min: 1194.93 / Avg: 1217.49 / Max: 1228.78Min: 1148.76 / Avg: 1171.03 / Max: 1205.51Min: 1163.48 / Avg: 1249.06 / Max: 1316.521. (F9X) gfortran options: -lopenblas -lFoX_dom -lFoX_sax -lFoX_wxml -lFoX_common -lFoX_utils -lFoX_fsys -lfftw3 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent -levent_pthreads -lutil -lm -lrt -lz

RELION

RELION - REgularised LIkelihood OptimisatioN - is a stand-alone computer program for Maximum A Posteriori refinement of (multiple) 3D reconstructions or 2D class averages in cryo-electron microscopy (cryo-EM). It is developed in the research group of Sjors Scheres at the MRC Laboratory of Molecular Biology. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 3.1.1Test: Basic - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance80160240320400SE +/- 2.94, N = 9SE +/- 2.97, N = 9SE +/- 3.10, N = 9349.85348.29348.341. (CXX) g++ options: -fopenmp -std=c++0x -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgWatts, Fewer Is BetterRELION 3.1.1CPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance90180270360450Min: 120.6 / Avg: 480.05 / Max: 492.78Min: 119.68 / Avg: 480.3 / Max: 492.94Min: 119.71 / Avg: 480.54 / Max: 494.18
OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 3.1.1Test: Basic - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance60120180240300Min: 345.73 / Avg: 349.85 / Max: 373.26Min: 344.39 / Avg: 348.29 / Max: 371.97Min: 344.31 / Avg: 348.34 / Max: 373.091. (CXX) g++ options: -fopenmp -std=c++0x -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -pthread -lmpi_cxx -lmpi

BlogBench

BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: ReadLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance200K400K600K800K1000KSE +/- 10984.18, N = 9SE +/- 1738.41, N = 3SE +/- 13508.02, N = 31084405110311810863471. (CC) gcc options: -O2 -pthread
OpenBenchmarking.orgFinal Score Per Watt, More Is BetterBlogBench 1.1Test: ReadLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance100020003000400050004599.664752.604638.96
OpenBenchmarking.orgWatts, Fewer Is BetterBlogBench 1.1CPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance60120180240300Min: 111.85 / Avg: 235.76 / Max: 332.58Min: 120.02 / Avg: 232.11 / Max: 341.5Min: 120.83 / Avg: 234.18 / Max: 328.63
OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: ReadLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance200K400K600K800K1000KMin: 1006413 / Avg: 1084405.44 / Max: 1116895Min: 1099845 / Avg: 1103118.33 / Max: 1105770Min: 1066904 / Avg: 1086347.33 / Max: 11123131. (CC) gcc options: -O2 -pthread

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 0.26Backend: BLASLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance9001800270036004500SE +/- 50.84, N = 3SE +/- 49.90, N = 9SE +/- 17.79, N = 34106406141471. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.26Backend: BLASLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance36912159.699.729.72
MinAvgMaxLinux 5.11 Git120.76423.8492.33Linux 5.11 Patched120.35417.7492.34CPUFreq Performance121.03426.6492.36OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.26CPU Power Consumption Monitor130260390520650
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: BLASLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance7001400210028003500Min: 4010 / Avg: 4106 / Max: 4183Min: 3811 / Avg: 4060.89 / Max: 4359Min: 4119 / Avg: 4147 / Max: 41801. (CXX) g++ options: -flto -pthread

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance4K8K12K16K20KSE +/- 24.98, N = 3SE +/- 213.45, N = 3SE +/- 199.64, N = 91846817015173351. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops Per Watt, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance306090120150138.49127.83131.21
MinAvgMaxLinux 5.11 Git68.24133.4146.37Linux 5.11 Patched120.07133.1142.83CPUFreq Performance65.17132.1150.06OpenBenchmarking.orgWatts, Fewer Is BetterFFTW 3.3.6CPU Power Consumption Monitor4080120160200
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance3K6K9K12K15KMin: 18432 / Avg: 18468 / Max: 18516Min: 16653 / Avg: 17015.33 / Max: 17392Min: 16583 / Avg: 17334.89 / Max: 186651. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance510152025SE +/- 0.20, N = 15SE +/- 0.16, N = 15SE +/- 0.20, N = 1522.0922.4922.43
OpenBenchmarking.orgFPS Per Watt, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance0.0180.0360.0540.0720.090.080.080.08
OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLCPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance70140210280350Min: 121.59 / Avg: 288.37 / Max: 364.75Min: 120.81 / Avg: 287.74 / Max: 365.75Min: 121.23 / Avg: 291.09 / Max: 365.95
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance510152025Min: 20.88 / Avg: 22.09 / Max: 23.28Min: 21.2 / Avg: 22.49 / Max: 23.35Min: 20.74 / Avg: 22.43 / Max: 23.39

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance1.04852.0973.14554.1945.2425SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 34.654.634.66
OpenBenchmarking.orgFPS Per Watt, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance0.00680.01360.02040.02720.0340.030.030.03
OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLCPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance4080120160200Min: 78.18 / Avg: 157.61 / Max: 186.31Min: 77.57 / Avg: 156.39 / Max: 196.21Min: 82.95 / Avg: 160.37 / Max: 200.84
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance246810Min: 4.61 / Avg: 4.65 / Max: 4.68Min: 4.58 / Avg: 4.63 / Max: 4.67Min: 4.58 / Avg: 4.66 / Max: 4.71

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance612182430SE +/- 0.22, N = 15SE +/- 0.30, N = 15SE +/- 0.24, N = 1525.1225.4225.22
OpenBenchmarking.orgFPS Per Watt, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance0.02030.04060.06090.08120.10150.090.090.09
OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLCPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance60120180240300Min: 112.67 / Avg: 282.44 / Max: 341.96Min: 120.47 / Avg: 284.69 / Max: 345.62Min: 120.32 / Avg: 284.73 / Max: 343.46
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance612182430Min: 23.51 / Avg: 25.12 / Max: 26.29Min: 23.51 / Avg: 25.42 / Max: 27.36Min: 23.35 / Avg: 25.22 / Max: 26.66

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 0 - Input: 1080pLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance0.02070.04140.06210.08280.1035SE +/- 0.001, N = 3SE +/- 0.001, N = 12SE +/- 0.000, N = 30.0920.0910.0911. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 0 - Input: 1080pLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance
MinAvgMaxLinux 5.11 Git120.69154.1307.6Linux 5.11 Patched73.78154.7305.59CPUFreq Performance119.66152.9301.34OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 0.8CPU Power Consumption Monitor80160240320400
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 0 - Input: 1080pLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance12345Min: 0.09 / Avg: 0.09 / Max: 0.09Min: 0.09 / Avg: 0.09 / Max: 0.09Min: 0.09 / Avg: 0.09 / Max: 0.091. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

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 0.26Backend: EigenLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance10002000300040005000SE +/- 49.20, N = 4SE +/- 36.23, N = 3SE +/- 26.71, N = 34284443344501. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.26Backend: EigenLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance369121510.0810.4510.39
OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.26CPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance90180270360450Min: 122 / Avg: 424.99 / Max: 492.34Min: 121.42 / Avg: 424.24 / Max: 492.36Min: 121.85 / Avg: 428.45 / Max: 492.33
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: EigenLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance8001600240032004000Min: 4171 / Avg: 4283.5 / Max: 4398Min: 4385 / Avg: 4433 / Max: 4504Min: 4409 / Avg: 4449.67 / Max: 45001. (CXX) g++ options: -flto -pthread

AI Benchmark Alpha

AI Benchmark Alpha is a Python library for evaluating artificial intelligence (AI) performance on diverse hardware platforms and relies upon the TensorFlow machine learning library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device AI ScoreLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance6001200180024003000275627872908
OpenBenchmarking.orgScore Per Watt, More Is BetterAI Benchmark Alpha 0.1.2Device AI ScoreLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance369121512.9512.9811.59
MinAvgMaxLinux 5.11 Git120.43212.9368.5Linux 5.11 Patched119.91214.7378.22CPUFreq Performance119.81250.8388.86OpenBenchmarking.orgWatts, Fewer Is BetterAI Benchmark Alpha 0.1.2CPU Power Consumption Monitor100200300400500

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device Inference ScoreLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance400800120016002000169717201775

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device Training ScoreLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance2004006008001000105910671133

KeyDB

A benchmark of KeyDB as a multi-threaded fork of the Redis server. The KeyDB benchmark is conducted using memtier-benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance60K120K180K240K300KSE +/- 4239.68, N = 15SE +/- 3012.50, N = 15SE +/- 5131.54, N = 15302893.56294214.37303171.331. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.orgOps/sec Per Watt, More Is BetterKeyDB 6.0.16Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance4008001200160020001929.561865.161761.51
MinAvgMaxLinux 5.11 Git120.76157.0179.41Linux 5.11 Patched64.37157.7178.64CPUFreq Performance60.8172.1190.06OpenBenchmarking.orgWatts, Fewer Is BetterKeyDB 6.0.16CPU Power Consumption Monitor50100150200250
OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance50K100K150K200K250KMin: 275189.6 / Avg: 302893.56 / Max: 338020.92Min: 273233.06 / Avg: 294214.37 / Max: 314258.83Min: 272657.95 / Avg: 303171.33 / Max: 333275.991. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance612182430SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 324.9925.0825.121. (CXX) g++ options: -O3 -pthread -lm
OpenBenchmarking.orgns/day Per Watt, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance0.01350.0270.04050.0540.06750.060.060.06
MinAvgMaxLinux 5.11 Git120.36427.7438.57Linux 5.11 Patched119.89428.6441.65CPUFreq Performance119.98427.6439.32OpenBenchmarking.orgWatts, Fewer Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020CPU Power Consumption Monitor120240360480600
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance612182430Min: 24.86 / Avg: 24.99 / Max: 25.17Min: 25.01 / Avg: 25.08 / Max: 25.16Min: 25.05 / Avg: 25.12 / Max: 25.151. (CXX) g++ options: -O3 -pthread -lm

BRL-CAD

BRL-CAD 7.28.0 is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.30.8VGR Performance MetricLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance140K280K420K560K700K6389716365216357251. (CXX) g++ options: -std=c++11 -pipe -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -rdynamic -lSM -lICE -lXi -lGLU -lGL -lGLdispatch -lX11 -lXext -lXrender -lpthread -ldl -luuid -lm
OpenBenchmarking.orgVGR Performance Metric Per Watt, More Is BetterBRL-CAD 7.30.8VGR Performance MetricLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance300600900120015001355.951369.511355.13
MinAvgMaxLinux 5.11 Git120.87471.2492.28Linux 5.11 Patched119.9464.8492.27CPUFreq Performance120.43469.1492.3OpenBenchmarking.orgWatts, Fewer Is BetterBRL-CAD 7.30.8CPU Power Consumption Monitor130260390520650

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: super-resolution-10 - Device: OpenMP CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance9001800270036004500SE +/- 78.33, N = 9SE +/- 44.10, N = 3SE +/- 68.07, N = 124393421041901. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.6Model: super-resolution-10 - Device: OpenMP CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance4812162015.7215.2615.49
MinAvgMaxLinux 5.11 Git61.08279.5301.53Linux 5.11 Patched121.22275.8301.54CPUFreq Performance121.3270.4293.8OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.6CPU Power Consumption Monitor80160240320400
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: super-resolution-10 - Device: OpenMP CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance8001600240032004000Min: 4142 / Avg: 4393 / Max: 4922Min: 4127 / Avg: 4210.33 / Max: 4277Min: 3763 / Avg: 4189.54 / Max: 45001. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

YafaRay

YafaRay is an open-source physically based montecarlo ray-tracing engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterYafaRay 3.4.1Total Time For Sample SceneLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance20406080100SE +/- 0.38, N = 3SE +/- 0.85, N = 15SE +/- 0.69, N = 1584.7487.1487.331. (CXX) g++ options: -std=c++11 -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype -lpthread
MinAvgMaxLinux 5.11 Git122.2313.2371.24Linux 5.11 Patched77.37312.5377.49CPUFreq Performance120.76313.3374.38OpenBenchmarking.orgWatts, Fewer Is BetterYafaRay 3.4.1CPU Power Consumption Monitor100200300400500
OpenBenchmarking.orgSeconds, Fewer Is BetterYafaRay 3.4.1Total Time For Sample SceneLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance20406080100Min: 84.16 / Avg: 84.74 / Max: 85.45Min: 82.11 / Avg: 87.14 / Max: 91.42Min: 83.68 / Avg: 87.33 / Max: 90.791. (CXX) g++ options: -std=c++11 -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype -lpthread

Hierarchical INTegration

This test runs the U.S. Department of Energy's Ames Laboratory Hierarchical INTegration (HINT) benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQUIPs, More Is BetterHierarchical INTegration 1.0Test: FLOATLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance70M140M210M280M350MSE +/- 121683.80, N = 3SE +/- 122621.60, N = 3SE +/- 201175.83, N = 3322702844.18323144417.17322775283.801. (CC) gcc options: -O3 -march=native -lm
OpenBenchmarking.orgQUIPs Per Watt, More Is BetterHierarchical INTegration 1.0Test: FLOATLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance500K1000K1500K2000K2500K2518730.772496420.852527557.17
MinAvgMaxLinux 5.11 Git120.47128.1130.19Linux 5.11 Patched64.77129.4131.58CPUFreq Performance64.16127.7129.43OpenBenchmarking.orgWatts, Fewer Is BetterHierarchical INTegration 1.0CPU Power Consumption Monitor4080120160200
OpenBenchmarking.orgQUIPs, More Is BetterHierarchical INTegration 1.0Test: FLOATLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance60M120M180M240M300MMin: 322525956.21 / Avg: 322702844.18 / Max: 322936042.52Min: 322905080.69 / Avg: 323144417.17 / Max: 323310417.96Min: 322378619.67 / Avg: 322775283.8 / Max: 323031996.681. (CC) gcc options: -O3 -march=native -lm

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance40K80K120K160K200KSE +/- 7366.43, N = 15SE +/- 2393.85, N = 15SE +/- 2146.20, N = 15189771134044132844
MinAvgMaxLinux 5.11 Git123.16385.7435.87Linux 5.11 Patched121.52403.7446.41CPUFreq Performance121.76401.7439.74OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2020-08-23CPU Power Consumption Monitor120240360480600
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance30K60K90K120K150KMin: 138046 / Avg: 189770.67 / Max: 231817Min: 106838 / Avg: 134043.8 / Max: 146394Min: 119861 / Avg: 132844.47 / Max: 152016

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance30060090012001500SE +/- 26.23, N = 15SE +/- 3.00, N = 3SE +/- 35.45, N = 151317.401123.321247.48MIN: 1147.91MIN: 1077.56MIN: 1069.011. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 2.0CPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance90180270360450Min: 122.37 / Avg: 333.59 / Max: 492.01Min: 120.62 / Avg: 341.54 / Max: 492.22Min: 120.77 / Avg: 339.16 / Max: 492.27
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance2004006008001000Min: 1230.95 / Avg: 1317.4 / Max: 1634.02Min: 1118.58 / Avg: 1123.32 / Max: 1128.89Min: 1102.05 / Avg: 1247.48 / Max: 1519.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 BenchmarkLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance70140210280350SE +/- 1.74, N = 3SE +/- 0.23, N = 3SE +/- 0.25, N = 3323.93323.00324.48
OpenBenchmarking.orgScore Per Watt, More Is BetterNumpy BenchmarkLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance0.55351.1071.66052.2142.76752.422.432.46
MinAvgMaxLinux 5.11 Git67.38133.8152.2Linux 5.11 Patched120.11133.2142.03CPUFreq Performance119.8131.7141.04OpenBenchmarking.orgWatts, Fewer Is BetterNumpy BenchmarkCPU Power Consumption Monitor4080120160200
OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance60120180240300Min: 321.81 / Avg: 323.93 / Max: 327.39Min: 322.57 / Avg: 323 / Max: 323.37Min: 324.07 / Avg: 324.48 / Max: 324.94

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance4080120160200SE +/- 1.60, N = 12SE +/- 1.86, N = 3SE +/- 2.62, N = 31751811851. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance0.15750.3150.47250.630.78750.670.690.70
OpenBenchmarking.orgWatts, Fewer Is BetterONNX Runtime 1.6CPU Power Consumption MonitorLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance50100150200250Min: 122.51 / Avg: 262.95 / Max: 279.9Min: 121.06 / Avg: 263.88 / Max: 279.86Min: 121.25 / Avg: 263.54 / Max: 278.66
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPULinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance306090120150Min: 166 / Avg: 174.58 / Max: 184.5Min: 177.5 / Avg: 181.17 / Max: 183.5Min: 181 / Avg: 185.17 / Max: 1901. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

Gcrypt Library

Libgcrypt is a general purpose cryptographic library developed as part of the GnuPG project. This is a benchmark of libgcrypt's integrated benchmark and is measuring the time to run the benchmark command with a cipher/mac/hash repetition count set for 50 times as simple, high level look at the overall crypto performance of the system under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGcrypt Library 1.9Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance50100150200250SE +/- 0.85, N = 3SE +/- 0.81, N = 3SE +/- 0.62, N = 3233.83232.54232.761. (CC) gcc options: -O2 -fvisibility=hidden -lgpg-error
MinAvgMaxLinux 5.11 Git120.42129.2134.47Linux 5.11 Patched119.81129.2133.08CPUFreq Performance63.27127.6131.63OpenBenchmarking.orgWatts, Fewer Is BetterGcrypt Library 1.9CPU Power Consumption Monitor4080120160200
OpenBenchmarking.orgSeconds, Fewer Is BetterGcrypt Library 1.9Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance4080120160200Min: 232.33 / Avg: 233.83 / Max: 235.28Min: 231.18 / Avg: 232.54 / Max: 233.99Min: 232.1 / Avg: 232.76 / Max: 233.991. (CC) gcc options: -O2 -fvisibility=hidden -lgpg-error

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance50100150200250SE +/- 1.36, N = 3SE +/- 0.79, N = 3SE +/- 2.78, N = 3210.35208.79208.04
MinAvgMaxLinux 5.11 Git121.43319.1492.32Linux 5.11 Patched64.92318.0492.35CPUFreq Performance121.13322.9492.29OpenBenchmarking.orgWatts, Fewer Is BetterTimed LLVM Compilation 10.0CPU Power Consumption Monitor130260390520650
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance4080120160200Min: 208.65 / Avg: 210.35 / Max: 213.03Min: 207.68 / Avg: 208.79 / Max: 210.33Min: 203.35 / Avg: 208.04 / Max: 212.98

OpenFOAM

OpenFOAM is the leading free, open source software for computational fluid dynamics (CFD). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 60MLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance306090120150SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3129.67128.28128.811. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
MinAvgMaxLinux 5.11 Git122.11288.9399.64Linux 5.11 Patched121.03290.6400.88CPUFreq Performance121.09290.0399.67OpenBenchmarking.orgWatts, Fewer Is BetterOpenFOAM 8CPU Power Consumption Monitor110220330440550
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 60MLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance20406080100Min: 129.52 / Avg: 129.67 / Max: 129.84Min: 128.2 / Avg: 128.28 / Max: 128.42Min: 128.69 / Avg: 128.81 / Max: 128.991. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance10K20K30K40K50KSE +/- 759.19, N = 15SE +/- 400.94, N = 6SE +/- 211.03, N = 345083.941034.041180.1
MinAvgMaxLinux 5.11 Git123.66447.4480.16Linux 5.11 Patched121.22453.6483.06CPUFreq Performance121.59453.0481.15OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2020-08-23CPU Power Consumption Monitor120240360480600
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance8K16K24K32K40KMin: 42204.7 / Avg: 45083.93 / Max: 52593.3Min: 40070.1 / Avg: 41033.95 / Max: 42901.4Min: 40933.4 / Avg: 41180.07 / Max: 41600

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: CPU-OnlyLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance4080120160200SE +/- 0.14, N = 3SE +/- 0.13, N = 3SE +/- 0.95, N = 3158.64156.83158.07
MinAvgMaxLinux 5.11 Git122.22434.4492.49Linux 5.11 Patched120.64423.5492.26CPUFreq Performance121.24434.6493.96OpenBenchmarking.orgWatts, Fewer Is BetterBlender 2.90CPU Power Consumption Monitor130260390520650
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: CPU-OnlyLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance306090120150Min: 158.38 / Avg: 158.64 / Max: 158.83Min: 156.6 / Avg: 156.83 / Max: 157.05Min: 156.44 / Avg: 158.07 / Max: 159.73

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 3.1Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance714212835SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.01, N = 330.2430.8330.931. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance0.02030.04060.06090.08120.10150.080.080.09
MinAvgMaxLinux 5.11 Git121.53362.6451.49Linux 5.11 Patched120.45363.8391.83CPUFreq Performance119.84363.8399.1OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption Monitor120240360480600
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Linux 5.11 GitLinux 5.11 PatchedCPUFreq Performance714212835Min: 30.05 / Avg: 30.24 / Max: 30.47Min: 30.57 / Avg: 30.83 / Max: 30.95Min: 30.91 / Avg: 30.93 / Max: 30.941. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance14K28K42K56K70KSE +/- 690.93, N = 3SE +/- 412.91, N = 15SE +/- 715.70, N = 465193.062195.461347.2
MinAvgMaxLinux 5.11 Git123.35441.7476.85Linux 5.11 Patched121.35449.5481.23CPUFreq Performance121.89451.3481.68OpenBenchmarking.orgWatts, Fewer Is BetterTensorFlow Lite 2020-08-23CPU Power Consumption Monitor120240360480600
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance11K22K33K44K55KMin: 63856.6 / Avg: 65192.97 / Max: 66165.7Min: 59468.9 / Avg: 62195.4 / Max: 65210.3Min: 60282.3 / Avg: 61347.15 / Max: 63426.1

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: GarlicoinLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance2K4K6K8K10KSE +/- 88.01, N = 14SE +/- 99.38, N = 15SE +/- 155.17, N = 149937.319949.889964.351. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s Per Watt, More Is BetterCpuminer-Opt 3.15.5Algorithm: GarlicoinLinux 5.11 GitLinux 5.11 PatchedCPUFreq Performance142842567061.0659.9157.99
MinAvgMaxLinux 5.11 Git120.87162.7392.21Linux 5.11 Patched75.92166.1473.33CPUFreq Performance120.49171.8441.12OpenBenchmarking.orgWatts, Fewer Is BetterCpuminer-Opt 3.15.5CPU Power Consumption Monitor120240360480600