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 2102048-HA-AMDEPYC7F37
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 2 Tests
BLAS (Basic Linear Algebra Sub-Routine) Tests 3 Tests
C++ Boost Tests 2 Tests
C/C++ Compiler Tests 5 Tests
CPU Massive 12 Tests
Creator Workloads 6 Tests
Database Test Suite 2 Tests
Encoding 3 Tests
Fortran Tests 3 Tests
HPC - High Performance Computing 11 Tests
LAPACK (Linear Algebra Pack) Tests 2 Tests
Machine Learning 5 Tests
Molecular Dynamics 2 Tests
MPI Benchmarks 3 Tests
Multi-Core 11 Tests
NVIDIA GPU Compute 2 Tests
Intel oneAPI 2 Tests
OpenMPI Tests 6 Tests
Python Tests 2 Tests
Renderers 2 Tests
Scientific Computing 4 Tests
Server 2 Tests
Server CPU Tests 8 Tests
Video Encoding 3 Tests
Common Workstation Benchmarks 2 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Additional Graphs

Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

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

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
CPUFreq Performance
January 24 2021
  6 Hours, 39 Minutes
Linux 5.11 Git
January 22 2021
  5 Hours, 35 Minutes
Linux 5.11 Giovanni Patch
January 23 2021
  5 Hours, 41 Minutes
Linux 5.11 Rafael Patch
February 03 2021
  6 Hours, 39 Minutes
Invert Hiding All Results Option
  6 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):


AMD EPYC 7F72 2P Linux 5.11 Perf GovernorProcessorMotherboardChipsetMemoryDiskGraphicsNetworkMonitorOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen ResolutionCPUFreq PerformanceLinux 5.11 GitLinux 5.11 Giovanni PatchLinux 5.11 Rafael Patch2 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-rc4-max-boost-inv-patch (x86_64) 20210121GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.9GCC 10.2.0ext41920x10805.11.0-051100rc4daily20210122-generic (x86_64) 20210121VE2285.11.0-rc4-max-boost-inv-patch (x86_64) 202101212 x AMD EPYC 7F72 24-Core @ 3.71GHz (48 Cores / 96 Threads)5.11.0-rc6-phx (x86_64) 20210203aspeedOpenBenchmarking.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- CPUFreq Performance, Linux 5.11 Git, Linux 5.11 Giovanni Patch: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Details- CPUFreq Performance: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034- Linux 5.11 Git: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301034- Linux 5.11 Giovanni Patch: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301034- Linux 5.11 Rafael Patch: Scaling Governor: acpi-cpufreq schedutil (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

CPUFreq PerformanceLinux 5.11 GitLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchResult OverviewPhoronix Test Suite100%111%123%134%Cpuminer-Optdav1dx265OSPrayLeelaChessZeroLAMMPS Molecular Dynamics SimulatorTimed GDB GNU Debugger CompilationInfluxDBDaCapo BenchmarkCLOMPRodiniaQMCPACKrav1eQuantum ESPRESSOoneDNNTTSIOD 3D RendererONNX RuntimeAI Benchmark AlphaRedisNAS Parallel BenchmarksOpenFOAMTensorFlow Lite

AMD EPYC 7F72 2P Linux 5.11 Perf Governordav1d: Summer Nature 4Kdav1d: Chimera 1080p 10-bitospray: Magnetic Reconnection - Path Tracerospray: San Miguel - SciVisttsiod-renderer: Phong Rendering With Soft-Shadow Mappingx265: Bosphorus 1080prav1e: 10x265: Bosphorus 4Krav1e: 6rav1e: 5onnx: yolov4 - OpenMP CPUcpuminer-opt: Skeincoincpuminer-opt: LBC, LBRY Creditslczero: BLASlczero: Eigenlammps: Rhodopsin Proteinredis: SETai-benchmark: Device Inference Scoreai-benchmark: Device Training Scoreai-benchmark: Device AI Scoreclomp: Static OMP Speedupnpb: LU.Cinfluxdb: 64 - 10000 - 2,5000,1 - 10000influxdb: 4 - 10000 - 2,5000,1 - 10000tensorflow-lite: Mobilenet Floattensorflow-lite: Mobilenet Quanttensorflow-lite: NASNet Mobiletensorflow-lite: SqueezeNettensorflow-lite: Inception ResNet V2tensorflow-lite: Inception V4onednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUdacapobench: Jythondacapobench: Tradebeansdacapobench: Tradesoapbuild-gdb: Time To Compileopenfoam: Motorbike 30Mopenfoam: Motorbike 60Mqe: AUSURF112rodinia: OpenMP CFD Solverrodinia: OpenMP Streamclusterqmcpack: simple-H2OCPUFreq PerformanceLinux 5.11 GitLinux 5.11 Giovanni PatchLinux 5.11 Rafael Patch363.33181.95333.3355.56665.44862.293.17720.751.4461.0951855229481940874147445024.6391454741.4217751133290847.4153770.571360163.0956189.439981.941180.113284461347.27379938188870.8675452.304670.8136280.51460447404671462185.27718.29128.811249.068.51010.43228.971308.29130.6125052.63627.20847.662.90218.631.3701.0451753637841324774106428421.1291380890.2216971059275643.9147443.861231991.2807463.146659.245083.918977165193.07657268946400.9141982.405490.8813480.54767448975954517097.64118.71129.671217.499.25511.20931.177317.45133.3725054.97655.22549.453.05419.741.4081.0681813640171390374061443323.7871427348.1017201067278747.8154376.761256112.1812193.639523.541034.013404462195.47362858107500.8637822.332900.8492480.52196847785591514892.91618.30128.281171.038.88210.33829.281315.99133.6925054.97642.25150.433.04519.931.3981.0641813666481375573522374924.7951401394.3017131061277446.9154111.431264917.7812464.340060.040544.113546762054.27244548154780.8703512.311710.8238250.51180948435600516793.09818.32129.051222.608.95210.56729.040OpenBenchmarking.org

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Summer Nature 4KCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git80160240320400SE +/- 3.54, N = 15SE +/- 0.53, N = 3SE +/- 2.68, N = 3SE +/- 1.86, N = 3363.33317.45315.99308.29MIN: 186.32 / MAX: 403.05MIN: 173.69 / MAX: 340.43MIN: 157.61 / MAX: 340.72MIN: 163.13 / MAX: 334.131. (CC) gcc options: -pthread

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Chimera 1080p 10-bitCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git4080120160200SE +/- 0.23, N = 3SE +/- 0.19, N = 3SE +/- 0.14, N = 3SE +/- 0.23, N = 3181.95133.69133.37130.61MIN: 125.32 / MAX: 275.36MIN: 92.24 / MAX: 206.31MIN: 92.59 / MAX: 205.11MIN: 90.23 / MAX: 199.741. (CC) gcc options: -pthread

OSPray

OpenBenchmarking.orgFPS Per Watt, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Git0.07430.14860.22290.29720.37150.330.320.30

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: Path TracerCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git70140210280350SE +/- 0.00, N = 11333.33250.00250.00250.00MIN: 100 / MAX: 500MIN: 111.11 / MAX: 333.33MIN: 90.91 / MAX: 333.33MIN: 90.91 / MAX: 500

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git1224364860SE +/- 0.00, N = 3SE +/- 0.58, N = 5SE +/- 0.58, N = 5SE +/- 0.00, N = 355.5654.9754.9752.63MIN: 33.33 / MAX: 58.82MIN: 25.64 / MAX: 58.82MIN: 31.25 / MAX: 58.82MIN: 27.03 / MAX: 58.82

TTSIOD 3D Renderer

A portable GPL 3D software renderer that supports OpenMP and Intel Threading Building Blocks with many different rendering modes. This version does not use OpenGL but is entirely CPU/software based. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterTTSIOD 3D Renderer 2.3bPhong Rendering With Soft-Shadow MappingCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git140280420560700SE +/- 5.92, N = 15SE +/- 3.22, N = 3SE +/- 24.62, N = 12SE +/- 9.04, N = 15665.45655.23642.25627.211. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++

rav1e

OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betterrav1e 0.4Speed: 10Linux 5.11 Giovanni PatchLinux 5.11 GitCPUFreq Performance0.00450.0090.01350.0180.02250.020.020.02

OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betterrav1e 0.4Speed: 6Linux 5.11 Giovanni PatchLinux 5.11 GitCPUFreq Performance0.00230.00460.00690.00920.01150.010.010.01

x265

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

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git1428425670SE +/- 0.73, N = 15SE +/- 0.14, N = 3SE +/- 0.52, N = 4SE +/- 0.42, N = 762.2950.4349.4547.661. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

rav1e

OpenBenchmarking.orgFrames Per Second Per Watt, More Is Betterrav1e 0.4Speed: 5Linux 5.11 Giovanni PatchLinux 5.11 GitCPUFreq Performance0.00230.00460.00690.00920.01150.010.010.01

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 10CPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git0.71481.42962.14442.85923.574SE +/- 0.018, N = 3SE +/- 0.008, N = 3SE +/- 0.011, N = 3SE +/- 0.016, N = 33.1773.0543.0452.902

x265

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

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git510152025SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.14, N = 3SE +/- 0.10, N = 320.7519.9319.7418.631. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 6CPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git0.32540.65080.97621.30161.627SE +/- 0.001, N = 3SE +/- 0.003, N = 3SE +/- 0.002, N = 3SE +/- 0.002, N = 31.4461.4081.3981.370

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 5CPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git0.24640.49280.73920.98561.232SE +/- 0.003, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.001, N = 31.0951.0681.0641.045

ONNX Runtime

OpenBenchmarking.orgInferences Per Minute Per Watt, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPUCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Git0.15750.3150.47250.630.78750.700.690.67

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPUCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git4080120160200SE +/- 2.62, N = 3SE +/- 1.86, N = 4SE +/- 1.86, N = 3SE +/- 1.60, N = 121851811811751. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

Cpuminer-Opt

OpenBenchmarking.orgkH/s Per Watt, More Is BetterCpuminer-Opt 3.15.5Algorithm: SkeincoinCPUFreq PerformanceLinux 5.11 GitLinux 5.11 Giovanni Patch80016002400320040003625.652899.042897.78

OpenBenchmarking.orgkH/s Per Watt, More Is BetterCpuminer-Opt 3.15.5Algorithm: LBC, LBRY CreditsCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Git300600900120015001315.461091.621052.71

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: SkeincoinCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git110K220K330K440K550KSE +/- 9614.65, N = 12SE +/- 2614.77, N = 13SE +/- 5604.68, N = 12SE +/- 3597.13, N = 155229483666483640173637841. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: LBC, LBRY CreditsCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git40K80K120K160K200KSE +/- 861.90, N = 3SE +/- 1380.06, N = 3SE +/- 1545.00, N = 3SE +/- 1036.73, N = 31940871390371375571324771. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

LeelaChessZero

OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.26Backend: BLASLinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Git36912159.729.729.69

OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.26Backend: EigenLinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Git369121510.4510.3910.08

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: BLASCPUFreq PerformanceLinux 5.11 GitLinux 5.11 Giovanni PatchLinux 5.11 Rafael Patch9001800270036004500SE +/- 17.79, N = 3SE +/- 50.84, N = 3SE +/- 49.90, N = 9SE +/- 36.49, N = 541474106406135221. (CXX) g++ options: -flto -pthread

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: EigenCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 GitLinux 5.11 Rafael Patch10002000300040005000SE +/- 26.71, N = 3SE +/- 36.23, N = 3SE +/- 49.20, N = 4SE +/- 54.19, N = 944504433428437491. (CXX) g++ options: -flto -pthread

LAMMPS Molecular Dynamics Simulator

OpenBenchmarking.orgns/day Per Watt, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Git0.03150.0630.09450.1260.15750.140.130.12

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinLinux 5.11 Rafael PatchCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Git612182430SE +/- 0.16, N = 3SE +/- 0.19, N = 15SE +/- 0.17, N = 12SE +/- 0.23, N = 1524.8024.6423.7921.131. (CXX) g++ options: -O3 -pthread -lm

Redis

OpenBenchmarking.orgRequests Per Second Per Watt, More Is BetterRedis 6.0.9Test: SETCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Git3K6K9K12K15K11683.8511500.6111015.92

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SETCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git300K600K900K1200K1500KSE +/- 10017.79, N = 13SE +/- 13176.39, N = 15SE +/- 15335.21, N = 5SE +/- 10410.66, N = 151454741.421427348.101401394.301380890.221. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

AI Benchmark Alpha

OpenBenchmarking.orgScore Per Watt, More Is BetterAI Benchmark Alpha 0.1.2Device AI ScoreLinux 5.11 Giovanni PatchLinux 5.11 GitCPUFreq Performance369121512.9812.9511.59

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device Inference ScoreCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git4008001200160020001775172017131697

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device Training ScoreCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git20040060080010001133106710611059

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device AI ScoreCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git60012001800240030002908278727742756

CLOMP

CLOMP is the C version of the Livermore OpenMP benchmark developed to measure OpenMP overheads and other performance impacts due to threading in order to influence future system designs. This particular test profile configuration is currently set to look at the OpenMP static schedule speed-up across all available CPU cores using the recommended test configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP SpeedupLinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Git1122334455SE +/- 0.47, N = 3SE +/- 0.55, N = 3SE +/- 0.98, N = 15SE +/- 0.60, N = 347.847.446.943.91. (CC) gcc options: -fopenmp -O3 -lm

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.CLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchCPUFreq PerformanceLinux 5.11 Git30K60K90K120K150KSE +/- 509.59, N = 4SE +/- 237.90, N = 3SE +/- 121.33, N = 4SE +/- 1780.52, N = 15154376.76154111.43153770.57147443.861. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent -levent_pthreads -lutil -lm -lrt -lz 2. Open MPI 4.0.3

InfluxDB

OpenBenchmarking.orgval/sec Per Watt, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.11 GitCPUFreq PerformanceLinux 5.11 Giovanni Patch110022003300440055005060.304963.604857.31

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000CPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git300K600K900K1200K1500KSE +/- 9433.94, N = 3SE +/- 5312.25, N = 3SE +/- 2545.78, N = 3SE +/- 6204.63, N = 31360163.01264917.71256112.11231991.2

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000CPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git200K400K600K800K1000KSE +/- 2401.68, N = 3SE +/- 757.15, N = 3SE +/- 1525.09, N = 3SE +/- 2183.04, N = 3956189.4812464.3812193.6807463.1

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 FloatLinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Git10K20K30K40K50KSE +/- 395.37, N = 3SE +/- 473.02, N = 3SE +/- 126.68, N = 3SE +/- 1144.94, N = 1539523.539981.940060.046659.2

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Git10K20K30K40K50KSE +/- 60.30, N = 3SE +/- 400.94, N = 6SE +/- 211.03, N = 3SE +/- 759.19, N = 1540544.141034.041180.145083.9

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git40K80K120K160K200KSE +/- 2146.20, N = 15SE +/- 2393.85, N = 15SE +/- 1630.66, N = 3SE +/- 7366.43, N = 15132844134044135467189771

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git14K28K42K56K70KSE +/- 715.70, N = 4SE +/- 479.59, N = 15SE +/- 412.91, N = 15SE +/- 690.93, N = 361347.262054.262195.465193.0

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2Linux 5.11 Rafael PatchLinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Git160K320K480K640K800KSE +/- 1613.84, N = 3SE +/- 5824.36, N = 9SE +/- 2132.19, N = 3SE +/- 4257.59, N = 3724454736285737993765726

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4Linux 5.11 Giovanni PatchLinux 5.11 Rafael PatchCPUFreq PerformanceLinux 5.11 Git200K400K600K800K1000KSE +/- 1163.43, N = 3SE +/- 2213.60, N = 3SE +/- 4685.69, N = 3SE +/- 2435.29, N = 3810750815478818887894640

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: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPULinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Git0.20570.41140.61710.82281.0285SE +/- 0.001510, N = 7SE +/- 0.001697, N = 7SE +/- 0.007039, N = 3SE +/- 0.006064, N = 70.8637820.8675450.8703510.914198MIN: 0.79MIN: 0.78MIN: 0.8MIN: 0.781. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git0.54121.08241.62362.16482.706SE +/- 0.02560, N = 15SE +/- 0.02377, N = 15SE +/- 0.01587, N = 3SE +/- 0.03372, N = 32.304672.311712.332902.40549MIN: 1.88MIN: 1.89MIN: 2MIN: 1.921. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git0.19830.39660.59490.79320.9915SE +/- 0.005456, N = 5SE +/- 0.006790, N = 15SE +/- 0.004000, N = 5SE +/- 0.005127, N = 50.8136280.8238250.8492480.881348MIN: 0.69MIN: 0.68MIN: 0.73MIN: 0.711. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPULinux 5.11 Rafael PatchCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Git0.12320.24640.36960.49280.616SE +/- 0.003992, N = 10SE +/- 0.005906, N = 4SE +/- 0.004601, N = 4SE +/- 0.005010, N = 40.5118090.5146040.5219680.547674MIN: 0.43MIN: 0.43MIN: 0.43MIN: 0.431. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git10002000300040005000SE +/- 20.83, N = 6SE +/- 43.93, N = 6SE +/- 24.16, N = 4SE +/- 28.66, N = 184740477848434897

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git13002600390052006500SE +/- 52.34, N = 20SE +/- 66.39, N = 20SE +/- 53.96, N = 20SE +/- 50.83, N = 204671559156005954

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradesoapCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git11002200330044005500SE +/- 42.72, N = 5SE +/- 61.21, N = 4SE +/- 47.07, N = 4SE +/- 44.82, N = 44621514851675170

Timed GDB GNU Debugger Compilation

This test times how long it takes to build the GNU Debugger (GDB) in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 9.1Time To CompileCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git20406080100SE +/- 0.14, N = 3SE +/- 0.43, N = 3SE +/- 0.23, N = 3SE +/- 0.40, N = 385.2892.9293.1097.64

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 30MCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git510152025SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.16, N = 3SE +/- 0.14, N = 318.2918.3018.3218.711. (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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 60MLinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Git306090120150SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3128.28128.81129.05129.671. (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

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 Giovanni PatchLinux 5.11 GitLinux 5.11 Rafael PatchCPUFreq Performance30060090012001500SE +/- 12.21, N = 4SE +/- 11.28, N = 3SE +/- 23.97, N = 6SE +/- 19.05, N = 91171.031217.491222.601249.061. (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

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 CFD SolverCPUFreq PerformanceLinux 5.11 Giovanni PatchLinux 5.11 Rafael PatchLinux 5.11 Git3691215SE +/- 0.053, N = 5SE +/- 0.141, N = 15SE +/- 0.168, N = 15SE +/- 0.148, N = 158.5108.8828.9529.2551. (CXX) g++ options: -O2 -lOpenCL

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP StreamclusterLinux 5.11 Giovanni PatchCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Git3691215SE +/- 0.03, N = 5SE +/- 0.04, N = 5SE +/- 0.13, N = 4SE +/- 0.22, N = 1510.3410.4310.5711.211. (CXX) g++ options: -O2 -lOpenCL

QMCPACK

QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. QMCPACK is an open-source production level many-body ab initio Quantum Monte Carlo code for computing the electronic structure of atoms, molecules, and solids. QMCPACK is supported by the U.S. Department of Energy. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.10Input: simple-H2OCPUFreq PerformanceLinux 5.11 Rafael PatchLinux 5.11 Giovanni PatchLinux 5.11 Git714212835SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.53, N = 1528.9729.0429.2831.181. (CXX) g++ options: -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -fomit-frame-pointer -ffast-math -pthread -lm