Xeon Scalable Ice Lake P-State Governor

Intel Xeon Platinum 8380 Ice Lake P-State CPU frequency scaling Linux benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2105267-IB-XEONSCALA38
Jump To Table - Results

View

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

Limit displaying results to tests within:

AV1 3 Tests
BLAS (Basic Linear Algebra Sub-Routine) Tests 4 Tests
C++ Boost Tests 2 Tests
Timed Code Compilation 5 Tests
C/C++ Compiler Tests 10 Tests
CPU Massive 14 Tests
Creator Workloads 12 Tests
Encoding 6 Tests
Fortran Tests 7 Tests
Game Development 4 Tests
HPC - High Performance Computing 16 Tests
LAPACK (Linear Algebra Pack) Tests 4 Tests
Machine Learning 2 Tests
Molecular Dynamics 6 Tests
MPI Benchmarks 5 Tests
Multi-Core 26 Tests
NVIDIA GPU Compute 4 Tests
Intel oneAPI 4 Tests
OpenMPI Tests 11 Tests
Programmer / Developer System Benchmarks 7 Tests
Python Tests 5 Tests
Renderers 2 Tests
Scientific Computing 8 Tests
Server CPU Tests 11 Tests
Video Encoding 6 Tests
Common Workstation Benchmarks 2 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
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
Performance Per
Dollar
Date
Run
  Test
  Duration
P-State powersave
May 23 2021
  15 Hours, 42 Minutes
P-State performance
May 24 2021
  11 Hours, 14 Minutes
intel_cpufreq schedutil
May 24 2021
  15 Hours, 54 Minutes
intel_cpufreq performance
May 25 2021
  13 Hours, 43 Minutes
Invert Hiding All Results Option
  14 Hours, 8 Minutes

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


Xeon Scalable Ice Lake P-State GovernorOpenBenchmarking.orgPhoronix Test Suite2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads)Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS)Intel Device 099816 x 32 GB DDR4-3200MT/s Hynix HMA84GR7CJR4N-XN800GB INTEL SSDPF21Q800GBASPEED2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFPUbuntu 21.045.11.0-17-generic (x86_64)GNOME Shell 3.38.4X Server1.0.2GCC 10.3.0ext41024x768ProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerVulkanCompilerFile-SystemScreen ResolutionXeon Scalable Ice Lake P-State Governor BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/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-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 - P-State powersave: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270 - P-State performance: Scaling Governor: intel_pstate performance - CPU Microcode: 0xd000270 - intel_cpufreq schedutil: Scaling Governor: intel_cpufreq schedutil - CPU Microcode: 0xd000270 - intel_cpufreq performance: Scaling Governor: intel_cpufreq performance - CPU Microcode: 0xd000270 - Python 3.9.4- 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

P-State powersaveP-State performanceintel_cpufreq schedutilintel_cpufreq performanceResult OverviewPhoronix Test Suite100%148%195%243%290%AOM AV1KvazaarPostgreSQL pgbenchSVT-VP9SVT-HEVCSVT-AV1PJSIPTimed Godot Game Engine CompilationDarmstadt Automotive Parallel Heterogeneous SuiteZstd CompressionArrayFireOpenVKLlibavif avifencTimed Linux Kernel CompilationTimed Node.js CompilationRodiniaTensorFlow LiteNAS Parallel BenchmarksQuantum ESPRESSOEmbreeOSPrayBlenderTimed LLVM CompilationIntel Open Image DenoiseTimed Mesa CompilationOpenFOAMONNX RuntimeHigh Performance Conjugate GradientStockfishGROMACSLAMMPS Molecular Dynamics SimulatorLiquid-DSPXcompact3d Incompact3dRELIONNAMDWRFNWChem

P-State powersaveP-State performanceintel_cpufreq schedutilintel_cpufreq performancePer Watt Result OverviewPhoronix Test Suite100%107%113%120%126%AOM AV1NAS Parallel BenchmarksSVT-HEVCIntel Open Image DenoiseOSPrayZstd CompressionKvazaarOpenVKLSVT-AV1StockfishEmbreeArrayFireDarmstadt Automotive Parallel Heterogeneous SuitePJSIPSVT-VP9ONNX RuntimeLiquid-DSPHigh Performance Conjugate GradientGROMACSLAMMPS Molecular Dynamics SimulatorP.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

Xeon Scalable Ice Lake P-State Governorwrf: conus 2.5kmqe: AUSURF112openvkl: vklBenchmarkUnstructuredVolumenwchem: C240 Buckyballopenvkl: vklBenchmarkonnx: super-resolution-10 - OpenMP CPUpgbench: 1000 - 250 - Read Only - Average Latencypgbench: 1000 - 250 - Read Onlydaphne: OpenMP - Points2Imageonnx: bertsquad-10 - OpenMP CPUrelion: Basic - CPUaom-av1: Speed 6 Two-Pass - Bosphorus 4Kincompact3d: X3D-benchmarking input.i3dlammps: 20k Atomsrodinia: OpenMP Leukocyteonnx: shufflenet-v2-10 - OpenMP CPUpgbench: 1000 - 100 - Read Only - Average Latencypgbench: 1000 - 100 - Read Onlyopenvkl: vklBenchmarkVdbVolumepjsip: INVITEbuild-llvm: Unix Makefileshpcg: openfoam: Motorbike 60Mcompress-zstd: 19 - Decompression Speedcompress-zstd: 19 - Compression Speedbuild-llvm: Ninjaonnx: fcn-resnet101-11 - OpenMP CPUonnx: yolov4 - OpenMP CPUblender: Barbershop - CPU-Onlytensorflow-lite: Mobilenet Quantrodinia: OpenMP HotSpot3Dpgbench: 1000 - 250 - Read Write - Average Latencypgbench: 1000 - 250 - Read Writepgbench: 1000 - 100 - Read Write - Average Latencypgbench: 1000 - 100 - Read Writecompress-zstd: 19, Long Mode - Decompression Speedcompress-zstd: 19, Long Mode - Compression Speedbuild-nodejs: Time To Compilepjsip: OPTIONS, Statelessdaphne: OpenMP - NDT Mappingbuild-linux-kernel: Time To Compiledaphne: OpenMP - Euclidean Clusterblender: Classroom - CPU-Onlyavifenc: 6, Losslesspjsip: OPTIONS, Statefulaom-av1: Speed 6 Realtime - Bosphorus 4Ktensorflow-lite: Inception ResNet V2tensorflow-lite: Inception V4openfoam: Motorbike 30Mbuild-godot: Time To Compileaom-av1: Speed 9 Realtime - Bosphorus 4Kopenvkl: vklBenchmarkStructuredVolumestockfish: Total Timeospray: San Miguel - Path Tracerrodinia: OpenMP LavaMDcompress-zstd: 8, Long Mode - Compression Speedkvazaar: Bosphorus 4K - Very Fastliquid-dsp: 64 - 256 - 57kvazaar: Bosphorus 4K - Ultra Fastnpb: EP.Dgromacs: MPI CPU - water_GMX50_bareblender: BMW27 - CPU-Onlyavifenc: 10, Losslessaom-av1: Speed 8 Realtime - Bosphorus 4Koidn: RT.hdr_alb_nrm.3840x2160rodinia: OpenMP Streamclusterospray: San Miguel - SciVisospray: XFrog Forest - Path Tracersvt-av1: Preset 8 - Bosphorus 4Knamd: ATPase Simulation - 327,506 Atomsoidn: RTLightmap.hdr.4096x4096embree: Pathtracer ISPC - Asian Dragon Objnpb: BT.Cospray: XFrog Forest - SciVisliquid-dsp: 160 - 256 - 57liquid-dsp: 128 - 256 - 57arrayfire: Conjugate Gradient CPUnpb: SP.Cnpb: IS.Darrayfire: BLAS CPUavifenc: 10build-mesa: Time To Compileoidn: RT.ldr_alb_nrm.3840x2160ospray: NASA Streamlines - Path Tracerembree: Pathtracer ISPC - Crownnpb: LU.Cincompact3d: input.i3d 193 Cells Per Directionrodinia: OpenMP CFD Solversvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080pembree: Pathtracer ISPC - Asian Dragonsvt-vp9: VMAF Optimized - Bosphorus 1080psvt-hevc: 7 - Bosphorus 1080pospray: Magnetic Reconnection - SciVisnpb: FT.Cnpb: CG.Cospray: NASA Streamlines - SciVisnpb: SP.Bnpb: EP.Csvt-hevc: 10 - Bosphorus 1080pnpb: MG.CP-State powersaveP-State performanceintel_cpufreq schedutilintel_cpufreq performance9892.421106.6917572691882.367068980.2779064346634.58509350.2663.94290.41110235.49569.04982120.256393329217223262548199.39840.1359105.032563.282.7129.272196480107.7341126.8104.33910.723234114.283233832711.646.7102.19741421742.3024.881831.3372.1737.46937697.3857206568215314.9275.61223.597197443817703112210.4239.815296.013.26302990000023.038620.978.99929.728.83317.072.857.71090.9110.3830.5460.270931.3887.2092191166.4118.87307393333332701333334.29291313.602999.724538.695.20518.9832.8727.7866.5251174911.0211.19165664.827206.54106.6219210.92168.38111.1190989.7339485.72125114946.466051.90258.65116754.169885.5561178.5218038401888.281166260.27192419013527.238996430477349.9939.69290.14333135.82546.35883920.0811236713299965395393190.68540.6471104.762561.886.7127.321194483105.1033656.3104.1853.774664931.588630492677.949.190.85042389924.8421.4861020.7771.2730.3451036219.0957214966370714.7948.15761.0610605645817713142810.4539.4231011.938.21306613333346.318920.109.05228.407.24848.622.937.81610010.3156.8870.271681.4089.0895197815.3118.87308510000032930000006.13191817.763014.774539.994.34118.4972.9327.7874.0551187300.4711.05545094.746453.67107.6307460.44305.77114.81100543.7240261.26142.86123544.097610.18576.99120084.279889.9071156.491748017188071867560.2858853797077.76504349.4364.00290.41669735.79263.28683340.0811228570235992492540191.17340.0659105.102594.781.5127.688194483107.3935239.0104.1843.814657262.740365772693.446.191.85741743743.0122.190884.7572.2431.98737975.8657163267225415.0753.00121.727799762817544618210.3839.199311.711.30300312500020.538634.309.04229.258.48615.902.787.71290.9110.3135.4800.272271.3887.9927191705.0218.87308253333332914666676.45491642.332926.264510.295.13418.7542.8827.7867.7540177878.6811.20541794.710260.25102.7043262.78193.80111.1191970.9939430.66125115501.416157.58345.48117869.499875.2641175.6918096671882.282169550.28887646213249.678695875503349.02810.01289.97558635.84947.10583800.0811231983300163185416191.80140.8311104.272609.783.1127.332197476104.5133581.3104.0883.791661861.594628262713.148.391.08142338914.7021.4101021.6271.2230.0141040919.2057267967742114.4948.65161.2710670464117862947410.4239.7461015.838.26307260000046.348938.099.10228.287.27648.912.937.72710010.3156.7500.270511.4089.4905197821.8818.87309670000032934000004.00591830.562978.524574.834.34018.5172.9327.7874.2420186973.0611.05617484.729457.02106.1736464.71309.13112.96100763.1140226.02142.86123176.727705.44574.88120124.05OpenBenchmarking.org

WRF

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWRF 4.2.2Input: conus 2.5kmP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil2K4K6K8K10K9885.569892.429875.269889.911. (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 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz

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: AUSURF112P-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil30060090012001500SE +/- 1.74, N = 3SE +/- 19.34, N = 9SE +/- 17.96, N = 9SE +/- 22.92, N = 91178.521106.691175.691156.491. (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_core -levent_pthreads -lutil -lm -lrt -lz

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkUnstructuredVolumeP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil400K800K1200K1600K2000KSE +/- 8888.13, N = 3SE +/- 19689.28, N = 3SE +/- 4355.52, N = 3SE +/- 4960.83, N = 31803840175726918096671748017MIN: 24013 / MAX: 5810552MIN: 18102 / MAX: 5872318MIN: 24362 / MAX: 5849692MIN: 13879 / MAX: 5825120

NWChem

NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil4008001200160020001888.21882.31882.21880.01. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lz -lcomex -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil2004006008001000SE +/- 2.67, N = 3SE +/- 2.19, N = 3SE +/- 2.40, N = 3SE +/- 8.72, N = 3811670821718MIN: 1 / MAX: 3166MIN: 1 / MAX: 2830MIN: 1 / MAX: 3226MIN: 1 / MAX: 3134

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 CPUP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil15003000450060007500SE +/- 200.91, N = 12SE +/- 205.17, N = 12SE +/- 156.45, N = 12SE +/- 210.11, N = 1266266898695567561. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

PostgreSQL pgbench

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1000 - Clients: 250 - Mode: Read Only - Average LatencyP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil0.06480.12960.19440.25920.324SE +/- 0.003, N = 3SE +/- 0.002, N = 15SE +/- 0.006, N = 15SE +/- 0.005, N = 150.2710.2770.2880.2851. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1000 - Clients: 250 - Mode: Read OnlyP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil200K400K600K800K1000KSE +/- 9754.80, N = 3SE +/- 7359.52, N = 15SE +/- 16646.22, N = 15SE +/- 15445.86, N = 159241909064348764628853791. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Darmstadt Automotive Parallel Heterogeneous Suite

DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTest Cases Per Minute, More Is BetterDarmstadt Automotive Parallel Heterogeneous SuiteBackend: OpenMP - Kernel: Points2ImageP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil3K6K9K12K15KSE +/- 114.27, N = 15SE +/- 86.30, N = 3SE +/- 100.42, N = 15SE +/- 115.67, N = 1213527.246634.5813249.687077.761. (CXX) g++ options: -O3 -std=c++11 -fopenmp

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: bertsquad-10 - Device: OpenMP CPUP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil110220330440550SE +/- 2.17, N = 3SE +/- 10.98, N = 9SE +/- 6.15, N = 12SE +/- 8.53, N = 124775095035041. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

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: CPUP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil80160240320400SE +/- 1.34, N = 3SE +/- 1.16, N = 3SE +/- 1.39, N = 3SE +/- 1.38, N = 3349.99350.27349.03349.441. (CXX) g++ options: -fopenmp -std=c++0x -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -pthread -lmpi_cxx -lmpi

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4KP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil3691215SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 129.693.9410.014.001. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Xcompact3d Incompact3d

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

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: X3D-benchmarking input.i3dP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil60120180240300SE +/- 0.01, N = 3SE +/- 0.12, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 3290.14290.41289.98290.421. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz

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 AtomsP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil816243240SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 335.8335.5035.8535.791. (CXX) g++ options: -O3 -pthread -lm

Rodinia

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

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LeukocyteP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil1530456075SE +/- 0.64, N = 12SE +/- 0.69, N = 6SE +/- 0.69, N = 15SE +/- 1.35, N = 1546.3669.0547.1163.291. (CXX) g++ options: -O2 -lOpenCL

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: shufflenet-v2-10 - Device: OpenMP CPUP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil2K4K6K8K10KSE +/- 25.12, N = 3SE +/- 58.56, N = 12SE +/- 24.06, N = 3SE +/- 65.66, N = 383928212838083341. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

PostgreSQL pgbench

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1000 - Clients: 100 - Mode: Read Only - Average LatencyP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil0.05760.11520.17280.23040.288SE +/- 0.000, N = 3SE +/- 0.005, N = 15SE +/- 0.000, N = 3SE +/- 0.001, N = 30.0810.2560.0810.0811. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1000 - Clients: 100 - Mode: Read OnlyP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil300K600K900K1200K1500KSE +/- 5304.66, N = 3SE +/- 8388.30, N = 15SE +/- 3615.78, N = 3SE +/- 11834.71, N = 31236713393329123198312285701. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkVdbVolumeP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil6M12M18M24M30MSE +/- 268896.33, N = 15SE +/- 47580.68, N = 3SE +/- 349026.46, N = 15SE +/- 38895.86, N = 329996539217223263001631823599249MIN: 979600 / MAX: 247803840MIN: 1031810 / MAX: 151043112MIN: 979586 / MAX: 253907928MIN: 1060641 / MAX: 156949416

PJSIP

PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITEP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil12002400360048006000SE +/- 7.17, N = 3SE +/- 29.86, N = 15SE +/- 15.21, N = 3SE +/- 50.72, N = 1553932548541625401. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

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 12.0Build System: Unix MakefilesP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil4080120160200SE +/- 1.07, N = 3SE +/- 1.18, N = 3SE +/- 0.59, N = 3SE +/- 1.14, N = 3190.69199.40191.80191.17

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 3.1P-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil918273645SE +/- 0.20, N = 3SE +/- 0.17, N = 3SE +/- 0.21, N = 3SE +/- 0.12, N = 340.6540.1440.8340.071. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

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 60MP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil20406080100SE +/- 0.09, N = 3SE +/- 0.19, N = 3SE +/- 0.16, N = 3SE +/- 0.38, N = 3104.76105.03104.27105.101. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -ldecompose -lgenericPatchFields -lmetisDecomp -lscotchDecomp -llagrangian -lregionModels -lOpenFOAM -ldl -lm

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression SpeedP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil6001200180024003000SE +/- 5.29, N = 15SE +/- 16.96, N = 3SE +/- 4.37, N = 3SE +/- 4.60, N = 152561.82563.22609.72594.71. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression SpeedP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil20406080100SE +/- 0.75, N = 15SE +/- 0.55, N = 3SE +/- 0.32, N = 3SE +/- 0.75, N = 1586.782.783.181.51. (CC) gcc options: -O3 -pthread -lz -llzma

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 12.0Build System: NinjaP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil306090120150SE +/- 0.50, N = 3SE +/- 0.40, N = 3SE +/- 0.45, N = 3SE +/- 0.14, N = 3127.32129.27127.33127.69

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: fcn-resnet101-11 - Device: OpenMP CPUP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil4080120160200SE +/- 0.44, N = 3SE +/- 2.36, N = 3SE +/- 0.29, N = 3SE +/- 1.89, N = 31941961971941. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPUP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil100200300400500SE +/- 5.43, N = 3SE +/- 2.02, N = 3SE +/- 0.73, N = 3SE +/- 5.29, N = 34834804764831. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CPU-OnlyP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil20406080100SE +/- 0.34, N = 3SE +/- 0.44, N = 3SE +/- 0.20, N = 3SE +/- 0.55, N = 3105.10107.73104.51107.39

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 QuantP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil9K18K27K36K45KSE +/- 117.93, N = 3SE +/- 711.79, N = 12SE +/- 61.65, N = 3SE +/- 352.77, N = 333656.341126.833581.335239.0

Rodinia

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

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP HotSpot3DP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil20406080100SE +/- 0.01, N = 3SE +/- 0.44, N = 3SE +/- 0.52, N = 3SE +/- 0.27, N = 3104.19104.34104.09104.181. (CXX) g++ options: -O2 -lOpenCL

PostgreSQL pgbench

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1000 - Clients: 250 - Mode: Read Write - Average LatencyP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil3691215SE +/- 0.023, N = 3SE +/- 0.071, N = 3SE +/- 0.023, N = 3SE +/- 0.017, N = 33.77410.7233.7913.8141. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1000 - Clients: 250 - Mode: Read WriteP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil14K28K42K56K70KSE +/- 391.43, N = 3SE +/- 156.62, N = 3SE +/- 405.25, N = 3SE +/- 301.39, N = 3664932341166186657261. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1000 - Clients: 100 - Mode: Read Write - Average LatencyP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil0.96371.92742.89113.85484.8185SE +/- 0.007, N = 3SE +/- 0.009, N = 3SE +/- 0.004, N = 3SE +/- 0.035, N = 31.5884.2831.5942.7401. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1000 - Clients: 100 - Mode: Read WriteP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil14K28K42K56K70KSE +/- 272.63, N = 3SE +/- 51.95, N = 3SE +/- 148.39, N = 3SE +/- 464.84, N = 3630492338362826365771. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Decompression SpeedP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil6001200180024003000SE +/- 9.19, N = 3SE +/- 4.94, N = 15SE +/- 2.06, N = 3SE +/- 8.19, N = 62677.92711.62713.12693.41. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression SpeedP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil1122334455SE +/- 0.21, N = 3SE +/- 0.32, N = 15SE +/- 0.20, N = 3SE +/- 0.46, N = 649.146.748.346.11. (CC) gcc options: -O3 -pthread -lz -llzma

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 15.11Time To CompileP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil20406080100SE +/- 0.15, N = 3SE +/- 0.59, N = 3SE +/- 0.49, N = 3SE +/- 0.27, N = 390.85102.2091.0891.86

PJSIP

PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatelessP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil9K18K27K36K45KSE +/- 274.68, N = 3SE +/- 341.07, N = 9SE +/- 83.53, N = 3SE +/- 380.67, N = 7423894142142338417431. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

Darmstadt Automotive Parallel Heterogeneous Suite

DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTest Cases Per Minute, More Is BetterDarmstadt Automotive Parallel Heterogeneous SuiteBackend: OpenMP - Kernel: NDT MappingP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil2004006008001000SE +/- 4.79, N = 3SE +/- 18.61, N = 15SE +/- 4.97, N = 3SE +/- 19.99, N = 14924.84742.30914.70743.011. (CXX) g++ options: -O3 -std=c++11 -fopenmp

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. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil612182430SE +/- 0.24, N = 5SE +/- 0.38, N = 15SE +/- 0.19, N = 7SE +/- 0.19, N = 1421.4924.8821.4122.19

Darmstadt Automotive Parallel Heterogeneous Suite

DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTest Cases Per Minute, More Is BetterDarmstadt Automotive Parallel Heterogeneous SuiteBackend: OpenMP - Kernel: Euclidean ClusterP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil2004006008001000SE +/- 2.39, N = 3SE +/- 7.84, N = 15SE +/- 1.25, N = 3SE +/- 9.84, N = 151020.77831.331021.62884.751. (CXX) g++ options: -O3 -std=c++11 -fopenmp

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-OnlyP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil1632486480SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.15, N = 3SE +/- 0.13, N = 371.2772.1771.2272.24

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil918273645SE +/- 0.22, N = 3SE +/- 0.28, N = 15SE +/- 0.20, N = 3SE +/- 0.41, N = 330.3537.4730.0131.991. (CXX) g++ options: -O3 -fPIC -lm

PJSIP

PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatefulP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil2K4K6K8K10KSE +/- 0.58, N = 3SE +/- 31.56, N = 3SE +/- 6.00, N = 3SE +/- 13.93, N = 31036237691040937971. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KP-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil510152025SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.23, N = 3SE +/- 0.02, N = 319.097.3819.205.861. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

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: Inception ResNet V2P-State performanceP-State powersaveintel_cpufreq performanceintel_cpufreq schedutil120K240K360K480K600KSE +/- 2262.00, N = 3SE +/- 1289.77, N = 3SE +/- 491.89, N = 3SE +/- 1875.50, N = 3572149572065572679571632