EPYC 7F72

AMD EPYC 7F72 24-Core testing with a Supermicro H11DSi-NT v2.00 (2.1 BIOS) and llvmpipe on Ubuntu 20.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2012129-HA-EPYC7F72896
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:

Chess Test Suite 3 Tests
Timed Code Compilation 2 Tests
C/C++ Compiler Tests 12 Tests
CPU Massive 19 Tests
Creator Workloads 13 Tests
Database Test Suite 5 Tests
Encoding 4 Tests
Fortran Tests 4 Tests
HPC - High Performance Computing 15 Tests
Imaging 3 Tests
Common Kernel Benchmarks 2 Tests
Machine Learning 8 Tests
Molecular Dynamics 3 Tests
MPI Benchmarks 4 Tests
Multi-Core 15 Tests
NVIDIA GPU Compute 3 Tests
OpenMPI Tests 4 Tests
Programmer / Developer System Benchmarks 3 Tests
Python 2 Tests
Scientific Computing 6 Tests
Server 6 Tests
Server CPU Tests 11 Tests
Single-Threaded 6 Tests
Speech 2 Tests
Telephony 2 Tests
Video Encoding 4 Tests
Common Workstation Benchmarks 2 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
EPYC 7F72
December 10 2020
  12 Hours, 13 Minutes
AMD 7F72
December 11 2020
  11 Hours, 50 Minutes
AMD EPYC 7F72
December 11 2020
  11 Hours, 39 Minutes
Invert Hiding All Results Option
  11 Hours, 54 Minutes

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

EPYC 7F72

Compiler Notes: --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 Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301034
Python Notes: Python 3.8.6
Security Notes: 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

AMD 7F72

AMD EPYC 7F72

Processor: AMD EPYC 7F72 24-Core @ 3.20GHz (24 Cores / 48 Threads), Motherboard: Supermicro H11DSi-NT v2.00 (2.1 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Western Digital WD_BLACK SN850 1TB, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel 10G X550T

OS: Ubuntu 20.10, Kernel: 5.8.0-29-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.5 Mesa 20.2.1 (LLVM 11.0.0 256 bits), Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080

EPYC 7F72OpenBenchmarking.orgPhoronix Test SuiteAMD EPYC 7F72 24-Core @ 3.20GHz (24 Cores / 48 Threads)Supermicro H11DSi-NT v2.00 (2.1 BIOS)AMD Starship/Matisse64GB1000GB Western Digital WD_BLACK SN850 1TBllvmpipeVE2282 x Intel 10G X550TUbuntu 20.105.8.0-29-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.94.5 Mesa 20.2.1 (LLVM 11.0.0 256 bits)GCC 10.2.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionEPYC 7F72 BenchmarksSystem Logs- --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 - NONE / errors=remount-ro,relatime,rw / Block Size: 4096- Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301034 - Python 3.8.6- 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

EPYC 7F72AMD 7F72AMD EPYC 7F72Result OverviewPhoronix Test Suite100%116%132%148%yquake2RedisHigh Performance Conjugate GradientTimed LLVM CompilationBRL-CADoneDNNFFTENAMDLAMMPS Molecular Dynamics SimulatorNumpy BenchmarkHPC ChallengeInfluxDBAI Benchmark AlphaStockfishasmFishNCNNMlpack BenchmarkPHPBenchHuginLZ4 CompressionCraftyx264GROMACSBasis UniversalKeyDBTimed Linux Kernel CompilationTimed Clash CompilationBYTE Unix BenchmarkLibRawHierarchical INTegrationrav1eLevelDBTNNKvazaareSpeak-NG Speech EngineWebP Image EncodePostgreSQL pgbenchRNNoiseIndigoBenchx265TensorFlow LiteTimed HMMer Search

EPYC 7F72yquake2: Software CPU - 1920 x 1080ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - mobilenetlammps: Rhodopsin Proteinncnn: CPU - squeezenethpcc: G-Rand Accessinfluxdb: 64 - 10000 - 2,5000,1 - 10000leveldb: Hot Readcompress-lz4: 3 - Compression Speedbuild-llvm: Time To Compilebrl-cad: VGR Performance Metriccompress-lz4: 9 - Compression Speedncnn: CPU - vgg16ncnn: CPU - efficientnet-b0ai-benchmark: Device Inference Scorencnn: CPU - googlenetonednn: Convolution Batch Shapes Auto - f32 - CPUmlpack: scikit_svmbasis: UASTC Level 0ncnn: CPU - resnet50ncnn: CPU - resnet18ffte: N=256, 3D Complex FFT Routinenamd: ATPase Simulation - 327,506 Atomsmlpack: scikit_linearridgeregressionncnn: CPU - yolov4-tinynumpy: leveldb: Seek Randhpcc: EP-DGEMMncnn: CPU - blazefaceai-benchmark: Device AI Scorestockfish: Total Timemlpack: scikit_qdaasmfish: 1024 Hash Memory, 26 Depthncnn: CPU - alexnetonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUleveldb: Rand Readpgbench: 1 - 50 - Read Onlyonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUpgbench: 100 - 50 - Read Only - Average Latencycompress-lz4: 3 - Decompression Speedpgbench: 1 - 1 - Read Write - Average Latencykvazaar: Bosphorus 1080p - Very Fastphpbench: PHP Benchmark Suitehpcc: Rand Ring Latencyonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUhugin: Panorama Photo Assistant + Stitching Timepgbench: 1 - 1 - Read Writeonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUrav1e: 1tensorflow-lite: NASNet Mobileonednn: IP Shapes 1D - u8s8f32 - CPUtnn: CPU - MobileNet v2onednn: Deconvolution Batch shapes_1d - f32 - CPUhpcc: Rand Ring Bandwidthonednn: Recurrent Neural Network Inference - u8s8f32 - CPUpgbench: 1 - 100 - Read Only - Average Latencypgbench: 100 - 50 - Read Onlyai-benchmark: Device Training Scorecompress-lz4: 9 - Decompression Speedcrafty: Elapsed Timewebp: Defaulttensorflow-lite: Mobilenet Floatx264: H.264 Video Encodinggromacs: Water Benchmarkx265: Bosphorus 1080ponednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUtensorflow-lite: Mobilenet Quantpgbench: 100 - 1 - Read Write - Average Latencykeydb: onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUleveldb: Fill Synconednn: Recurrent Neural Network Training - u8s8f32 - CPUpgbench: 1 - 100 - Read Onlyhpcc: G-HPLbuild-linux-kernel: Time To Compiletensorflow-lite: SqueezeNetbasis: UASTC Level 2kvazaar: Bosphorus 1080p - Mediumpgbench: 100 - 1 - Read Onlybyte: Dhrystone 2pgbench: 1 - 50 - Read Write - Average Latencybasis: ETC1Swebp: Quality 100, Lossless, Highest Compressionleveldb: Overwriteleveldb: Rand Fillrav1e: 10libraw: Post-Processing Benchmarkkvazaar: Bosphorus 4K - Ultra Fastwebp: Quality 100leveldb: Seq Fillpgbench: 1 - 50 - Read Writecompress-lz4: 1 - Compression Speedncnn: CPU-v2-v2 - mobilenet-v2basis: UASTC Level 3compress-lz4: 1 - Decompression Speedhint: FLOATleveldb: Rand Deleteonednn: Deconvolution Batch shapes_3d - f32 - CPUkvazaar: Bosphorus 4K - Mediumtensorflow-lite: Inception ResNet V2onednn: IP Shapes 1D - f32 - CPUkvazaar: Bosphorus 1080p - Ultra Fasttensorflow-lite: Inception V4pgbench: 100 - 1 - Read Writekvazaar: Bosphorus 1080p - Slowx265: Bosphorus 4Kkvazaar: Bosphorus 4K - Very Fastindigobench: CPU - Supercarpgbench: 100 - 100 - Read Onlykvazaar: Bosphorus 4K - Slowespeak: Text-To-Speech Synthesisleveldb: Rand Fillleveldb: Seq Fillpgbench: 1 - 1 - Read Onlylammps: 20k Atomspgbench: 100 - 100 - Read Writepgbench: 100 - 50 - Read Write - Average Latencypgbench: 100 - 100 - Read Write - Average Latencyinfluxdb: 4 - 10000 - 2,5000,1 - 10000rnnoise: pgbench: 100 - 50 - Read Writemlpack: scikit_icaleveldb: Overwritebasis: UASTC Level 2 + RDO Post-Processingonednn: Recurrent Neural Network Training - f32 - CPUbuild-clash: Time To Compilewebp: Quality 100, Losslesspgbench: 1 - 100 - Read Writewebp: Quality 100, Highest Compressionpgbench: 1 - 100 - Read Write - Average Latencyindigobench: CPU - Bedroomtnn: CPU - SqueezeNet v1.1rav1e: 5rav1e: 6hmmer: Pfam Database Searchpgbench: 100 - 100 - Read Only - Average Latencypgbench: 100 - 1 - Read Only - Average Latencypgbench: 1 - 50 - Read Only - Average Latencypgbench: 1 - 1 - Read Only - Average Latencyleveldb: Fill Syncncnn: CPU - mnasnetncnn: CPU - shufflenet-v2redis: SETredis: GETredis: LPUSHredis: SADDredis: LPOPonednn: IP Shapes 3D - u8s8f32 - CPUhpcc: Max Ping Pong Bandwidthhpcc: EP-STREAM Triadhpcc: G-Ptranshpcc: G-Fftehpcg: EPYC 7F72AMD 7F72AMD EPYC 7F7214.59.5221.4412.12921.260.031441339487.539.77250.78298.14533792449.7036.4812.29196022.072.8493624.457.57824.9513.82111181.736853100.881691.6131.05324.1364.33736.141004.0934735259754440.056434983710.386.3772539.8678472311.402352.778155.303200.09110606.30.49583.065687961.16869931.27550.00620191616.570.3491045171.43782295.1932.359462.71076934.6860.143551623151310630.974063091.61160165.3178.792.84760.612.3096161482.00.554424090.960.578226925.9761168.5801613.9169845287.2545038.86389588.116.90137.142433637455257.220.66349.36039.28023.123.33.03935.1141.992.59624.124219777.4310.0427.18211308.9321206770.05794209.8533.1504710.9511756631.73051142.441342877180536.2123.5324.0010.30952931810.7332.795228.038220.3272941915.789294862.3293.3951197999.821.1372147951.74229.121694.8691611.21462.28319.04221458.53846.6834.769275.5221.0371.385142.1150.1890.0410.0590.0344.59.5810.451495615.592038392.691295998.651658090.042134890.150.5928049566.4963.116277.757399.8723814.999523.79.6122.2611.70021.100.030511339130.140.99050.54289.65433239448.3336.5612.61200921.552.9034224.497.73324.5313.86111004.658736110.875321.5931.33321.8163.75136.765534.1335325342151739.496444874110.516.4139040.3838526121.386512.750045.359500.09010661.50.49382.465678971.16276924.02450.32520271606.750.3491036411.42645292.9012.341522.72515930.7110.143555320152310616.073829161.60259959.6177.862.83060.332.3063361260.70.556426284.570.575254930.7141163.4551610.9569877887.3107739.05489393.616.98036.972434537547272.920.57149.51139.17023.223.23.03235.0142.132.58524.024319740.1610.0627.28711271.4321880722.63689210.6513.1449910.9111799131.73388141.931347400180136.0923.5923.9810.33953058110.7032.759228.607220.8042939115.816295372.3243.3891199592.421.1262152151.82228.697693.8971613.70462.32319.01821448.54946.6944.775275.7031.0361.386142.1310.1890.0410.0590.0344.59.5110.351500640.311814383.801332610.561757206.721376509.230.83902510110.2593.382208.175538.4312114.973114.59.9422.2011.70521.870.030371297582.540.62349.31294.11334188548.9837.4812.54200722.012.8364825.027.60625.0314.10113220.197478080.892771.6230.75318.4764.87136.438804.1635275270402339.866356476610.526.3288440.0308420221.403462.782145.363300.09110548.80.49883.295735671.15717932.48950.46120091602.590.3461043621.42669294.6532.357262.73035928.1150.144551484152010685.773592171.61260331.6177.712.84660.252.2967261596.20.553424640.300.576657929.8861162.7141605.9069534486.8829738.90389812.716.92637.082444637624546.920.63049.57939.34223.223.23.02634.9642.172.59124.024249780.3810.0827.21011314.6322432634.13132210.3343.1387810.9211785901.72765142.031344893180736.1523.6023.9310.31452906610.7332.711228.317220.8672946215.779295522.3283.3881197198.021.1682148151.84228.708694.9951611.74461.62419.01521428.54846.7424.770275.3811.0371.385142.0800.1890.0410.0590.0344.59.5010.451475849.681897703.881317592.971723521.291379285.460.59170210456.2863.299238.285559.4189515.4451OpenBenchmarking.org

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: Software CPU - Resolution: 1920 x 1080AMD 7F72AMD EPYC 7F72EPYC 7F72612182430SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 323.714.514.51. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: Software CPU - Resolution: 1920 x 1080AMD 7F72AMD EPYC 7F72EPYC 7F72612182430Min: 23.6 / Avg: 23.73 / Max: 23.9Min: 14.5 / Avg: 14.53 / Max: 14.6Min: 14.5 / Avg: 14.5 / Max: 14.51. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU-v3-v3 - Model: mobilenet-v3EPYC 7F72AMD 7F72AMD EPYC 7F723691215SE +/- 0.20, N = 3SE +/- 0.11, N = 3SE +/- 0.09, N = 39.529.619.94MIN: 8.4 / MAX: 19.2MIN: 8.63 / MAX: 62.41MIN: 8.64 / MAX: 101.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU-v3-v3 - Model: mobilenet-v3EPYC 7F72AMD 7F72AMD EPYC 7F723691215Min: 9.2 / Avg: 9.52 / Max: 9.9Min: 9.46 / Avg: 9.61 / Max: 9.83Min: 9.84 / Avg: 9.94 / Max: 10.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mobilenetEPYC 7F72AMD EPYC 7F72AMD 7F72510152025SE +/- 0.66, N = 3SE +/- 0.56, N = 3SE +/- 0.49, N = 321.4422.2022.26MIN: 18.65 / MAX: 82.99MIN: 18.67 / MAX: 87.73MIN: 18.47 / MAX: 54.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mobilenetEPYC 7F72AMD EPYC 7F72AMD 7F72510152025Min: 20.75 / Avg: 21.44 / Max: 22.75Min: 21.24 / Avg: 22.2 / Max: 23.17Min: 21.3 / Avg: 22.26 / Max: 22.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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: Rhodopsin ProteinEPYC 7F72AMD EPYC 7F72AMD 7F723691215SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 312.1311.7111.701. (CXX) g++ options: -O3 -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinEPYC 7F72AMD EPYC 7F72AMD 7F7248121620Min: 11.96 / Avg: 12.13 / Max: 12.22Min: 11.67 / Avg: 11.71 / Max: 11.73Min: 11.55 / Avg: 11.7 / Max: 11.781. (CXX) g++ options: -O3 -pthread -lm

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: squeezenetAMD 7F72EPYC 7F72AMD EPYC 7F72510152025SE +/- 0.17, N = 3SE +/- 0.28, N = 3SE +/- 0.29, N = 321.1021.2621.87MIN: 18.23 / MAX: 105.85MIN: 18.08 / MAX: 111.42MIN: 18.11 / MAX: 113.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: squeezenetAMD 7F72EPYC 7F72AMD EPYC 7F72510152025Min: 20.8 / Avg: 21.1 / Max: 21.39Min: 20.89 / Avg: 21.26 / Max: 21.8Min: 21.31 / Avg: 21.87 / Max: 22.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Random AccessEPYC 7F72AMD 7F72AMD EPYC 7F720.00710.01420.02130.02840.0355SE +/- 0.00015, N = 3SE +/- 0.00105, N = 3SE +/- 0.00070, N = 30.031440.030510.030371. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Random AccessEPYC 7F72AMD 7F72AMD EPYC 7F7212345Min: 0.03 / Avg: 0.03 / Max: 0.03Min: 0.03 / Avg: 0.03 / Max: 0.03Min: 0.03 / Avg: 0.03 / Max: 0.031. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

InfluxDB

This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000EPYC 7F72AMD 7F72AMD EPYC 7F72300K600K900K1200K1500KSE +/- 1881.74, N = 3SE +/- 811.21, N = 3SE +/- 1608.37, N = 31339487.51339130.11297582.5
OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000EPYC 7F72AMD 7F72AMD EPYC 7F72200K400K600K800K1000KMin: 1337542.7 / Avg: 1339487.53 / Max: 1343250.3Min: 1337830.3 / Avg: 1339130.13 / Max: 1340620.9Min: 1295132.3 / Avg: 1297582.53 / Max: 1300612.6

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadEPYC 7F72AMD EPYC 7F72AMD 7F72918273645SE +/- 0.13, N = 3SE +/- 0.14, N = 3SE +/- 0.50, N = 439.7740.6240.991. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadEPYC 7F72AMD EPYC 7F72AMD 7F72918273645Min: 39.53 / Avg: 39.77 / Max: 39.97Min: 40.36 / Avg: 40.62 / Max: 40.85Min: 39.79 / Avg: 40.99 / Max: 42.091. (CXX) g++ options: -O3 -lsnappy -lpthread

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedEPYC 7F72AMD 7F72AMD EPYC 7F721122334455SE +/- 0.55, N = 4SE +/- 0.36, N = 3SE +/- 0.05, N = 350.7850.5449.311. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedEPYC 7F72AMD 7F72AMD EPYC 7F721020304050Min: 49.99 / Avg: 50.78 / Max: 52.35Min: 49.82 / Avg: 50.54 / Max: 50.94Min: 49.22 / Avg: 49.31 / Max: 49.361. (CC) gcc options: -O3

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 CompileAMD 7F72AMD EPYC 7F72EPYC 7F7260120180240300SE +/- 2.80, N = 3SE +/- 4.06, N = 3SE +/- 1.69, N = 3289.65294.11298.15
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileAMD 7F72AMD EPYC 7F72EPYC 7F7250100150200250Min: 285.32 / Avg: 289.65 / Max: 294.89Min: 287.13 / Avg: 294.11 / Max: 301.19Min: 295.53 / Avg: 298.14 / Max: 301.31

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 MetricAMD EPYC 7F72EPYC 7F72AMD 7F7270K140K210K280K350K3418853379243323941. (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

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedEPYC 7F72AMD EPYC 7F72AMD 7F721122334455SE +/- 0.51, N = 5SE +/- 0.42, N = 3SE +/- 0.54, N = 549.7048.9848.331. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedEPYC 7F72AMD EPYC 7F72AMD 7F721020304050Min: 48.53 / Avg: 49.7 / Max: 50.9Min: 48.14 / Avg: 48.98 / Max: 49.48Min: 46.91 / Avg: 48.33 / Max: 49.981. (CC) gcc options: -O3

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: vgg16EPYC 7F72AMD 7F72AMD EPYC 7F72918273645SE +/- 0.23, N = 3SE +/- 0.88, N = 3SE +/- 0.64, N = 336.4836.5637.48MIN: 33.2 / MAX: 130.64MIN: 33.32 / MAX: 133.5MIN: 33.78 / MAX: 126.411. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: vgg16EPYC 7F72AMD 7F72AMD EPYC 7F72816243240Min: 36.02 / Avg: 36.48 / Max: 36.8Min: 35.38 / Avg: 36.56 / Max: 38.29Min: 36.2 / Avg: 37.48 / Max: 38.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: efficientnet-b0EPYC 7F72AMD EPYC 7F72AMD 7F723691215SE +/- 0.25, N = 3SE +/- 0.12, N = 3SE +/- 0.09, N = 312.2912.5412.61MIN: 10.71 / MAX: 71.52MIN: 11.02 / MAX: 93.95MIN: 10.98 / MAX: 64.211. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: efficientnet-b0EPYC 7F72AMD EPYC 7F72AMD 7F7248121620Min: 11.79 / Avg: 12.29 / Max: 12.58Min: 12.35 / Avg: 12.54 / Max: 12.76Min: 12.45 / Avg: 12.61 / Max: 12.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 Inference ScoreAMD 7F72AMD EPYC 7F72EPYC 7F72400800120016002000200920071960

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: googlenetAMD 7F72AMD EPYC 7F72EPYC 7F72510152025SE +/- 0.20, N = 3SE +/- 0.41, N = 3SE +/- 0.24, N = 321.5522.0122.07MIN: 20.16 / MAX: 115.96MIN: 20.09 / MAX: 121.23MIN: 20.38 / MAX: 109.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: googlenetAMD 7F72AMD EPYC 7F72EPYC 7F72510152025Min: 21.27 / Avg: 21.55 / Max: 21.93Min: 21.43 / Avg: 22.01 / Max: 22.79Min: 21.71 / Avg: 22.07 / Max: 22.531. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUAMD EPYC 7F72EPYC 7F72AMD 7F720.65331.30661.95992.61323.2665SE +/- 0.01841, N = 3SE +/- 0.00847, N = 3SE +/- 0.00877, N = 32.836482.849362.90342MIN: 2.45MIN: 2.46MIN: 2.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUAMD EPYC 7F72EPYC 7F72AMD 7F72246810Min: 2.81 / Avg: 2.84 / Max: 2.87Min: 2.83 / Avg: 2.85 / Max: 2.86Min: 2.89 / Avg: 2.9 / Max: 2.921. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_svmEPYC 7F72AMD 7F72AMD EPYC 7F72612182430SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.15, N = 324.4524.4925.02
OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_svmEPYC 7F72AMD 7F72AMD EPYC 7F72612182430Min: 24.38 / Avg: 24.45 / Max: 24.51Min: 24.45 / Avg: 24.49 / Max: 24.52Min: 24.84 / Avg: 25.02 / Max: 25.31

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 0EPYC 7F72AMD EPYC 7F72AMD 7F72246810SE +/- 0.004, N = 3SE +/- 0.021, N = 3SE +/- 0.015, N = 37.5787.6067.7331. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 0EPYC 7F72AMD EPYC 7F72AMD 7F723691215Min: 7.57 / Avg: 7.58 / Max: 7.59Min: 7.57 / Avg: 7.61 / Max: 7.64Min: 7.71 / Avg: 7.73 / Max: 7.761. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet50AMD 7F72EPYC 7F72AMD EPYC 7F72612182430SE +/- 0.54, N = 3SE +/- 0.13, N = 3SE +/- 0.20, N = 324.5324.9525.03MIN: 22.55 / MAX: 91.3MIN: 22.93 / MAX: 118.48MIN: 23.18 / MAX: 107.41. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet50AMD 7F72EPYC 7F72AMD EPYC 7F72612182430Min: 23.71 / Avg: 24.53 / Max: 25.55Min: 24.71 / Avg: 24.95 / Max: 25.15Min: 24.83 / Avg: 25.03 / Max: 25.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet18EPYC 7F72AMD 7F72AMD EPYC 7F7248121620SE +/- 0.08, N = 3SE +/- 0.27, N = 3SE +/- 0.35, N = 313.8213.8614.10MIN: 12.13 / MAX: 30.86MIN: 11.95 / MAX: 34.76MIN: 12.23 / MAX: 36.591. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet18EPYC 7F72AMD 7F72AMD EPYC 7F7248121620Min: 13.68 / Avg: 13.82 / Max: 13.96Min: 13.46 / Avg: 13.86 / Max: 14.37Min: 13.41 / Avg: 14.1 / Max: 14.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

FFTE

FFTE is a package by Daisuke Takahashi to compute Discrete Fourier Transforms of 1-, 2- and 3- dimensional sequences of length (2^p)*(3^q)*(5^r). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 7.0N=256, 3D Complex FFT RoutineAMD EPYC 7F72EPYC 7F72AMD 7F7220K40K60K80K100KSE +/- 1183.24, N = 3SE +/- 1183.09, N = 3SE +/- 1197.97, N = 4113220.20111181.74111004.661. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 7.0N=256, 3D Complex FFT RoutineAMD EPYC 7F72EPYC 7F72AMD 7F7220K40K60K80K100KMin: 111881.55 / Avg: 113220.2 / Max: 115579.54Min: 109161.14 / Avg: 111181.74 / Max: 113258.33Min: 109385.83 / Avg: 111004.66 / Max: 114487.261. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp

NAMD

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

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsAMD 7F72EPYC 7F72AMD EPYC 7F720.20090.40180.60270.80361.0045SE +/- 0.00402, N = 3SE +/- 0.00562, N = 3SE +/- 0.01077, N = 30.875320.881690.89277
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsAMD 7F72EPYC 7F72AMD EPYC 7F72246810Min: 0.87 / Avg: 0.88 / Max: 0.88Min: 0.87 / Avg: 0.88 / Max: 0.89Min: 0.88 / Avg: 0.89 / Max: 0.91

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_linearridgeregressionAMD 7F72EPYC 7F72AMD EPYC 7F720.36450.7291.09351.4581.8225SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 31.591.611.62
OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_linearridgeregressionAMD 7F72EPYC 7F72AMD EPYC 7F72246810Min: 1.58 / Avg: 1.59 / Max: 1.6Min: 1.59 / Avg: 1.61 / Max: 1.62Min: 1.61 / Avg: 1.62 / Max: 1.64

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: yolov4-tinyAMD EPYC 7F72EPYC 7F72AMD 7F72714212835SE +/- 0.29, N = 3SE +/- 0.18, N = 3SE +/- 0.43, N = 330.7531.0531.33MIN: 28.27 / MAX: 134MIN: 28.99 / MAX: 116.2MIN: 28.52 / MAX: 71.51. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: yolov4-tinyAMD EPYC 7F72EPYC 7F72AMD 7F72714212835Min: 30.29 / Avg: 30.75 / Max: 31.28Min: 30.76 / Avg: 31.05 / Max: 31.39Min: 30.48 / Avg: 31.33 / Max: 31.871. (CXX) g++ options: -O3 -rdynamic -lgomp -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 BenchmarkEPYC 7F72AMD 7F72AMD EPYC 7F7270140210280350SE +/- 0.44, N = 3SE +/- 2.42, N = 3SE +/- 0.87, N = 3324.13321.81318.47
OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkEPYC 7F72AMD 7F72AMD EPYC 7F7260120180240300Min: 323.39 / Avg: 324.13 / Max: 324.92Min: 317.26 / Avg: 321.81 / Max: 325.51Min: 317.57 / Avg: 318.47 / Max: 320.21

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomAMD 7F72EPYC 7F72AMD EPYC 7F721428425670SE +/- 0.14, N = 3SE +/- 0.66, N = 3SE +/- 0.52, N = 363.7564.3464.871. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomAMD 7F72EPYC 7F72AMD EPYC 7F721326395265Min: 63.5 / Avg: 63.75 / Max: 63.98Min: 63.57 / Avg: 64.34 / Max: 65.65Min: 63.83 / Avg: 64.87 / Max: 65.41. (CXX) g++ options: -O3 -lsnappy -lpthread

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: EP-DGEMMAMD 7F72AMD EPYC 7F72EPYC 7F72816243240SE +/- 0.85, N = 3SE +/- 0.48, N = 3SE +/- 0.60, N = 336.7736.4436.141. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: EP-DGEMMAMD 7F72AMD EPYC 7F72EPYC 7F72816243240Min: 35.57 / Avg: 36.77 / Max: 38.41Min: 35.49 / Avg: 36.44 / Max: 37.09Min: 35.08 / Avg: 36.14 / Max: 37.161. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: blazefaceEPYC 7F72AMD 7F72AMD EPYC 7F720.9361.8722.8083.7444.68SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 34.094.134.16MIN: 3.49 / MAX: 10.46MIN: 3.53 / MAX: 17.37MIN: 3.55 / MAX: 15.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: blazefaceEPYC 7F72AMD 7F72AMD EPYC 7F72246810Min: 4.03 / Avg: 4.09 / Max: 4.16Min: 4.1 / Avg: 4.13 / Max: 4.17Min: 4.05 / Avg: 4.16 / Max: 4.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 ScoreAMD 7F72AMD EPYC 7F72EPYC 7F728001600240032004000353235273473

Stockfish

This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 12Total TimeAMD 7F72AMD EPYC 7F72EPYC 7F7211M22M33M44M55MSE +/- 678703.71, N = 3SE +/- 356097.54, N = 15SE +/- 458927.44, N = 155342151752704023525975441. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver
OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 12Total TimeAMD 7F72AMD EPYC 7F72EPYC 7F729M18M27M36M45MMin: 52076342 / Avg: 53421517.33 / Max: 54251565Min: 50673179 / Avg: 52704022.87 / Max: 55172057Min: 49156210 / Avg: 52597544.33 / Max: 558700721. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_qdaAMD 7F72AMD EPYC 7F72EPYC 7F72918273645SE +/- 0.18, N = 3SE +/- 0.31, N = 10SE +/- 0.08, N = 339.4939.8640.05
OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_qdaAMD 7F72AMD EPYC 7F72EPYC 7F72816243240Min: 39.3 / Avg: 39.49 / Max: 39.84Min: 39.28 / Avg: 39.86 / Max: 42.6Min: 39.89 / Avg: 40.05 / Max: 40.16

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthAMD 7F72EPYC 7F72AMD EPYC 7F7214M28M42M56M70MSE +/- 380311.53, N = 3SE +/- 282969.84, N = 3SE +/- 644607.06, N = 3644487416434983763564766
OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthAMD 7F72EPYC 7F72AMD EPYC 7F7211M22M33M44M55MMin: 63726239 / Avg: 64448740.67 / Max: 65015914Min: 64057538 / Avg: 64349837 / Max: 64915673Min: 62325986 / Avg: 63564766.33 / Max: 64493385

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: alexnetEPYC 7F72AMD 7F72AMD EPYC 7F723691215SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.18, N = 310.3810.5110.52MIN: 9.08 / MAX: 21.34MIN: 9.04 / MAX: 16.98MIN: 9.15 / MAX: 90.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: alexnetEPYC 7F72AMD 7F72AMD EPYC 7F723691215Min: 10.19 / Avg: 10.38 / Max: 10.49Min: 10.35 / Avg: 10.51 / Max: 10.62Min: 10.19 / Avg: 10.52 / Max: 10.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7F72EPYC 7F72AMD 7F72246810SE +/- 0.03710, N = 3SE +/- 0.00849, N = 3SE +/- 0.01141, N = 36.328846.377256.41390MIN: 5.76MIN: 5.79MIN: 5.781. (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: u8s8f32 - Engine: CPUAMD EPYC 7F72EPYC 7F72AMD 7F723691215Min: 6.26 / Avg: 6.33 / Max: 6.38Min: 6.37 / Avg: 6.38 / Max: 6.39Min: 6.39 / Avg: 6.41 / Max: 6.431. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadEPYC 7F72AMD EPYC 7F72AMD 7F72918273645SE +/- 0.18, N = 3SE +/- 0.21, N = 3SE +/- 0.44, N = 439.8740.0340.381. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadEPYC 7F72AMD EPYC 7F72AMD 7F72816243240Min: 39.63 / Avg: 39.87 / Max: 40.23Min: 39.66 / Avg: 40.03 / Max: 40.39Min: 39.65 / Avg: 40.38 / Max: 41.61. (CXX) g++ options: -O3 -lsnappy -lpthread

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read OnlyAMD 7F72EPYC 7F72AMD EPYC 7F72200K400K600K800K1000KSE +/- 4242.40, N = 3SE +/- 11183.06, N = 3SE +/- 9353.87, N = 38526128472318420221. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read OnlyAMD 7F72EPYC 7F72AMD EPYC 7F72150K300K450K600K750KMin: 847759.59 / Avg: 852612.47 / Max: 861066.44Min: 824865.85 / Avg: 847230.58 / Max: 858628.67Min: 828236.85 / Avg: 842022.29 / Max: 859867.441. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUAMD 7F72EPYC 7F72AMD EPYC 7F720.31580.63160.94741.26321.579SE +/- 0.00538, N = 3SE +/- 0.00911, N = 3SE +/- 0.00520, N = 31.386511.402351.40346MIN: 1.29MIN: 1.29MIN: 1.291. (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: u8s8f32 - Engine: CPUAMD 7F72EPYC 7F72AMD EPYC 7F72246810Min: 1.38 / Avg: 1.39 / Max: 1.4Min: 1.38 / Avg: 1.4 / Max: 1.41Min: 1.39 / Avg: 1.4 / Max: 1.411. (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: CPUAMD 7F72EPYC 7F72AMD EPYC 7F720.6261.2521.8782.5043.13SE +/- 0.01780, N = 3SE +/- 0.01168, N = 3SE +/- 0.01373, N = 32.750042.778152.78214MIN: 2.48MIN: 2.48MIN: 2.481. (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: CPUAMD 7F72EPYC 7F72AMD EPYC 7F72246810Min: 2.73 / Avg: 2.75 / Max: 2.79Min: 2.76 / Avg: 2.78 / Max: 2.8Min: 2.76 / Avg: 2.78 / Max: 2.811. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUEPYC 7F72AMD 7F72AMD EPYC 7F721.20672.41343.62014.82686.0335SE +/- 0.01527, N = 3SE +/- 0.02370, N = 3SE +/- 0.03533, N = 35.303205.359505.36330MIN: 4.97MIN: 4.96MIN: 4.961. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUEPYC 7F72AMD 7F72AMD EPYC 7F72246810Min: 5.27 / Avg: 5.3 / Max: 5.33Min: 5.31 / Avg: 5.36 / Max: 5.38Min: 5.3 / Avg: 5.36 / Max: 5.421. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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: 100 - Clients: 50 - Mode: Read Only - Average LatencyAMD 7F72EPYC 7F72AMD EPYC 7F720.02050.0410.06150.0820.1025SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.000, N = 30.0900.0910.0911. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average LatencyAMD 7F72EPYC 7F72AMD EPYC 7F7212345Min: 0.09 / Avg: 0.09 / Max: 0.09Min: 0.09 / Avg: 0.09 / Max: 0.09Min: 0.09 / Avg: 0.09 / Max: 0.091. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression SpeedAMD 7F72EPYC 7F72AMD EPYC 7F722K4K6K8K10KSE +/- 1.64, N = 3SE +/- 37.48, N = 4SE +/- 18.01, N = 310661.510606.310548.81. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression SpeedAMD 7F72EPYC 7F72AMD EPYC 7F722K4K6K8K10KMin: 10658.3 / Avg: 10661.47 / Max: 10663.8Min: 10520 / Avg: 10606.33 / Max: 10683.5Min: 10513.1 / Avg: 10548.77 / Max: 105711. (CC) gcc options: -O3

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: 1 - Clients: 1 - Mode: Read Write - Average LatencyAMD 7F72EPYC 7F72AMD EPYC 7F720.11210.22420.33630.44840.5605SE +/- 0.004, N = 3SE +/- 0.000, N = 3SE +/- 0.002, N = 30.4930.4950.4981. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average LatencyAMD 7F72EPYC 7F72AMD EPYC 7F72246810Min: 0.49 / Avg: 0.49 / Max: 0.5Min: 0.5 / Avg: 0.5 / Max: 0.5Min: 0.49 / Avg: 0.5 / Max: 0.51. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Very FastAMD EPYC 7F72EPYC 7F72AMD 7F7220406080100SE +/- 0.09, N = 3SE +/- 0.22, N = 3SE +/- 0.08, N = 383.2983.0682.461. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Very FastAMD EPYC 7F72EPYC 7F72AMD 7F721632486480Min: 83.2 / Avg: 83.29 / Max: 83.47Min: 82.8 / Avg: 83.06 / Max: 83.5Min: 82.32 / Avg: 82.46 / Max: 82.61. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteAMD EPYC 7F72EPYC 7F72AMD 7F72120K240K360K480K600KSE +/- 3310.62, N = 3SE +/- 5796.24, N = 3SE +/- 4364.31, N = 3573567568796567897
OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteAMD EPYC 7F72EPYC 7F72AMD 7F72100K200K300K400K500KMin: 567604 / Avg: 573567 / Max: 579041Min: 559904 / Avg: 568795.67 / Max: 579683Min: 562047 / Avg: 567896.67 / Max: 576432

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.5.0Test / Class: Random Ring LatencyAMD EPYC 7F72AMD 7F72EPYC 7F720.2630.5260.7891.0521.315SE +/- 0.01464, N = 3SE +/- 0.01886, N = 3SE +/- 0.00865, N = 31.157171.162761.168691. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.5.0Test / Class: Random Ring LatencyAMD EPYC 7F72AMD 7F72EPYC 7F72246810Min: 1.13 / Avg: 1.16 / Max: 1.18Min: 1.13 / Avg: 1.16 / Max: 1.2Min: 1.16 / Avg: 1.17 / Max: 1.191. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUAMD 7F72EPYC 7F72AMD EPYC 7F722004006008001000SE +/- 3.80, N = 3SE +/- 6.19, N = 3SE +/- 4.03, N = 3924.02931.28932.49MIN: 894.61MIN: 898.52MIN: 897.71. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUAMD 7F72EPYC 7F72AMD EPYC 7F72160320480640800Min: 920.19 / Avg: 924.02 / Max: 931.62Min: 921.25 / Avg: 931.27 / Max: 942.59Min: 925.34 / Avg: 932.49 / Max: 939.271. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Hugin

Hugin is an open-source, cross-platform panorama photo stitcher software package. This test profile times how long it takes to run the assistant and panorama photo stitching on a set of images. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHuginPanorama Photo Assistant + Stitching TimeEPYC 7F72AMD 7F72AMD EPYC 7F721122334455SE +/- 0.51, N = 3SE +/- 0.39, N = 3SE +/- 0.24, N = 350.0150.3350.46
OpenBenchmarking.orgSeconds, Fewer Is BetterHuginPanorama Photo Assistant + Stitching TimeEPYC 7F72AMD 7F72AMD EPYC 7F721020304050Min: 49.23 / Avg: 50.01 / Max: 50.96Min: 49.85 / Avg: 50.32 / Max: 51.09Min: 50.1 / Avg: 50.46 / Max: 50.92

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read WriteAMD 7F72EPYC 7F72AMD EPYC 7F72400800120016002000SE +/- 15.86, N = 3SE +/- 0.78, N = 3SE +/- 7.88, N = 32027201920091. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read WriteAMD 7F72EPYC 7F72AMD EPYC 7F72400800120016002000Min: 1995.18 / Avg: 2026.9 / Max: 2043.52Min: 2017.31 / Avg: 2018.73 / Max: 2020Min: 1995.66 / Avg: 2008.96 / Max: 2022.921. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F7230060090012001500SE +/- 2.00, N = 3SE +/- 2.70, N = 3SE +/- 3.74, N = 31602.591606.751616.57MIN: 1564.17MIN: 1564.52MIN: 1572.331. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F7230060090012001500Min: 1598.68 / Avg: 1602.59 / Max: 1605.27Min: 1602.3 / Avg: 1606.75 / Max: 1611.63Min: 1611.71 / Avg: 1616.57 / Max: 1623.921. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 0.4 AlphaSpeed: 1AMD 7F72EPYC 7F72AMD EPYC 7F720.07850.1570.23550.3140.3925SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 30.3490.3490.346
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 1AMD 7F72EPYC 7F72AMD EPYC 7F7212345Min: 0.35 / Avg: 0.35 / Max: 0.35Min: 0.35 / Avg: 0.35 / Max: 0.35Min: 0.34 / Avg: 0.35 / Max: 0.35

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 MobileAMD 7F72AMD EPYC 7F72EPYC 7F7220K40K60K80K100KSE +/- 239.98, N = 3SE +/- 60.45, N = 3SE +/- 165.43, N = 3103641104362104517
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileAMD 7F72AMD EPYC 7F72EPYC 7F7220K40K60K80K100KMin: 103196 / Avg: 103641.33 / Max: 104019Min: 104243 / Avg: 104361.67 / Max: 104441Min: 104228 / Avg: 104517 / Max: 104801

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUAMD 7F72AMD EPYC 7F72EPYC 7F720.32350.6470.97051.2941.6175SE +/- 0.00516, N = 3SE +/- 0.00056, N = 3SE +/- 0.00957, N = 31.426451.426691.43782MIN: 1.33MIN: 1.32MIN: 1.331. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUAMD 7F72AMD EPYC 7F72EPYC 7F72246810Min: 1.42 / Avg: 1.43 / Max: 1.44Min: 1.43 / Avg: 1.43 / Max: 1.43Min: 1.42 / Avg: 1.44 / Max: 1.451. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v2AMD 7F72AMD EPYC 7F72EPYC 7F7260120180240300SE +/- 0.39, N = 3SE +/- 0.88, N = 3SE +/- 0.66, N = 3292.90294.65295.19MIN: 281.47 / MAX: 324.04MIN: 281.79 / MAX: 332.39MIN: 281.63 / MAX: 334.081. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v2AMD 7F72AMD EPYC 7F72EPYC 7F7250100150200250Min: 292.14 / Avg: 292.9 / Max: 293.41Min: 292.94 / Avg: 294.65 / Max: 295.84Min: 294.14 / Avg: 295.19 / Max: 296.421. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUAMD 7F72AMD EPYC 7F72EPYC 7F720.53091.06181.59272.12362.6545SE +/- 0.02163, N = 3SE +/- 0.01986, N = 3SE +/- 0.00293, N = 32.341522.357262.35946MIN: 2.1MIN: 2.1MIN: 2.121. (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: CPUAMD 7F72AMD EPYC 7F72EPYC 7F72246810Min: 2.3 / Avg: 2.34 / Max: 2.38Min: 2.34 / Avg: 2.36 / Max: 2.4Min: 2.35 / Avg: 2.36 / Max: 2.361. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Random Ring BandwidthAMD EPYC 7F72AMD 7F72EPYC 7F720.61431.22861.84292.45723.0715SE +/- 0.03283, N = 3SE +/- 0.06549, N = 3SE +/- 0.01911, N = 32.730352.725152.710761. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Random Ring BandwidthAMD EPYC 7F72AMD 7F72EPYC 7F72246810Min: 2.66 / Avg: 2.73 / Max: 2.76Min: 2.6 / Avg: 2.73 / Max: 2.81Min: 2.67 / Avg: 2.71 / Max: 2.741. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F722004006008001000SE +/- 3.30, N = 3SE +/- 2.82, N = 3SE +/- 2.42, N = 3928.12930.71934.69MIN: 898.12MIN: 900MIN: 900.241. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F72160320480640800Min: 922.25 / Avg: 928.11 / Max: 933.67Min: 927.77 / Avg: 930.71 / Max: 936.35Min: 930.92 / Avg: 934.69 / Max: 939.191. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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: 1 - Clients: 100 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F720.03240.06480.09720.12960.162SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1430.1430.1441. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 100 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F7212345Min: 0.14 / Avg: 0.14 / Max: 0.14Min: 0.14 / Avg: 0.14 / Max: 0.14Min: 0.14 / Avg: 0.14 / Max: 0.141. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read OnlyAMD 7F72EPYC 7F72AMD EPYC 7F72120K240K360K480K600KSE +/- 2634.03, N = 3SE +/- 5773.06, N = 3SE +/- 1352.44, N = 35553205516235514841. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read OnlyAMD 7F72EPYC 7F72AMD EPYC 7F72100K200K300K400K500KMin: 550086.48 / Avg: 555320.19 / Max: 558457.14Min: 540076.77 / Avg: 551622.87 / Max: 557410.44Min: 549204.72 / Avg: 551483.89 / Max: 553884.941. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

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 Training ScoreAMD 7F72AMD EPYC 7F72EPYC 7F7230060090012001500152315201513

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedAMD EPYC 7F72EPYC 7F72AMD 7F722K4K6K8K10KSE +/- 8.87, N = 3SE +/- 28.09, N = 5SE +/- 22.43, N = 510685.710630.910616.01. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedAMD EPYC 7F72EPYC 7F72AMD 7F722K4K6K8K10KMin: 10668 / Avg: 10685.7 / Max: 10695.7Min: 10579.5 / Avg: 10630.88 / Max: 10710.9Min: 10577.3 / Avg: 10616.04 / Max: 106711. (CC) gcc options: -O3

Crafty

This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed TimeEPYC 7F72AMD 7F72AMD EPYC 7F721.6M3.2M4.8M6.4M8MSE +/- 14432.27, N = 3SE +/- 15235.97, N = 3SE +/- 62574.81, N = 37406309738291673592171. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed TimeEPYC 7F72AMD 7F72AMD EPYC 7F721.3M2.6M3.9M5.2M6.5MMin: 7383037 / Avg: 7406309.33 / Max: 7432733Min: 7361528 / Avg: 7382916.33 / Max: 7412407Min: 7235678 / Avg: 7359216.67 / Max: 74383201. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: DefaultAMD 7F72EPYC 7F72AMD EPYC 7F720.36270.72541.08811.45081.8135SE +/- 0.004, N = 3SE +/- 0.003, N = 3SE +/- 0.004, N = 31.6021.6111.6121. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: DefaultAMD 7F72EPYC 7F72AMD EPYC 7F72246810Min: 1.6 / Avg: 1.6 / Max: 1.61Min: 1.6 / Avg: 1.61 / Max: 1.62Min: 1.61 / Avg: 1.61 / Max: 1.621. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

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 FloatAMD 7F72EPYC 7F72AMD EPYC 7F7213K26K39K52K65KSE +/- 39.74, N = 3SE +/- 43.27, N = 3SE +/- 126.43, N = 359959.660165.360331.6
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatAMD 7F72EPYC 7F72AMD EPYC 7F7210K20K30K40K50KMin: 59883.2 / Avg: 59959.57 / Max: 60016.8Min: 60108.8 / Avg: 60165.27 / Max: 60250.3Min: 60082.9 / Avg: 60331.63 / Max: 60495.4

x264

This is a simple test of the x264 encoder run on the CPU (OpenCL support disabled) with a sample video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2019-12-17H.264 Video EncodingEPYC 7F72AMD 7F72AMD EPYC 7F724080120160200SE +/- 1.85, N = 3SE +/- 1.47, N = 3SE +/- 1.51, N = 3178.79177.86177.711. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2019-12-17H.264 Video EncodingEPYC 7F72AMD 7F72AMD EPYC 7F72306090120150Min: 175.25 / Avg: 178.79 / Max: 181.51Min: 174.93 / Avg: 177.86 / Max: 179.5Min: 174.73 / Avg: 177.71 / Max: 179.571. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing on the CPU with the water_GMX50 data. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkEPYC 7F72AMD EPYC 7F72AMD 7F720.64061.28121.92182.56243.203SE +/- 0.004, N = 3SE +/- 0.003, N = 3SE +/- 0.009, N = 32.8472.8462.8301. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkEPYC 7F72AMD EPYC 7F72AMD 7F72246810Min: 2.84 / Avg: 2.85 / Max: 2.85Min: 2.84 / Avg: 2.85 / Max: 2.85Min: 2.82 / Avg: 2.83 / Max: 2.851. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm

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 1080pEPYC 7F72AMD 7F72AMD EPYC 7F721428425670SE +/- 0.19, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 360.6160.3360.251. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pEPYC 7F72AMD 7F72AMD EPYC 7F721224364860Min: 60.35 / Avg: 60.61 / Max: 60.98Min: 60.24 / Avg: 60.33 / Max: 60.38Min: 60.09 / Avg: 60.25 / Max: 60.381. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F720.51971.03941.55912.07882.5985SE +/- 0.00103, N = 3SE +/- 0.00978, N = 3SE +/- 0.00115, N = 32.296722.306332.30961MIN: 2.18MIN: 2.18MIN: 2.181. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F72246810Min: 2.29 / Avg: 2.3 / Max: 2.3Min: 2.29 / Avg: 2.31 / Max: 2.32Min: 2.31 / Avg: 2.31 / Max: 2.311. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -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: Mobilenet QuantAMD 7F72EPYC 7F72AMD EPYC 7F7213K26K39K52K65KSE +/- 111.88, N = 3SE +/- 90.15, N = 3SE +/- 119.05, N = 361260.761482.061596.2
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantAMD 7F72EPYC 7F72AMD EPYC 7F7211K22K33K44K55KMin: 61071.2 / Avg: 61260.67 / Max: 61458.5Min: 61320.3 / Avg: 61482.03 / Max: 61631.9Min: 61383.4 / Avg: 61596.2 / Max: 61795.1

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: 100 - Clients: 1 - Mode: Read Write - Average LatencyAMD EPYC 7F72EPYC 7F72AMD 7F720.12510.25020.37530.50040.6255SE +/- 0.002, N = 3SE +/- 0.001, N = 3SE +/- 0.004, N = 30.5530.5540.5561. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Write - Average LatencyAMD EPYC 7F72EPYC 7F72AMD 7F72246810Min: 0.55 / Avg: 0.55 / Max: 0.56Min: 0.55 / Avg: 0.55 / Max: 0.56Min: 0.55 / Avg: 0.56 / Max: 0.561. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

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 6.0.16AMD 7F72AMD EPYC 7F72EPYC 7F7290K180K270K360K450KSE +/- 4696.92, N = 3SE +/- 2330.52, N = 3SE +/- 1068.73, N = 3426284.57424640.30424090.961. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16AMD 7F72AMD EPYC 7F72EPYC 7F7270K140K210K280K350KMin: 417825.88 / Avg: 426284.57 / Max: 434052.42Min: 421052.5 / Avg: 424640.3 / Max: 429010.97Min: 422532.63 / Avg: 424090.96 / Max: 426137.121. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUAMD 7F72AMD EPYC 7F72EPYC 7F720.13010.26020.39030.52040.6505SE +/- 0.002676, N = 3SE +/- 0.001616, N = 3SE +/- 0.001786, N = 30.5752540.5766570.578226MIN: 0.52MIN: 0.52MIN: 0.521. (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: CPUAMD 7F72AMD EPYC 7F72EPYC 7F72246810Min: 0.57 / Avg: 0.58 / Max: 0.58Min: 0.57 / Avg: 0.58 / Max: 0.58Min: 0.58 / Avg: 0.58 / Max: 0.581. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUEPYC 7F72AMD EPYC 7F72AMD 7F722004006008001000SE +/- 1.31, N = 3SE +/- 7.20, N = 3SE +/- 0.29, N = 3925.98929.89930.71MIN: 896.94MIN: 896.55MIN: 898.241. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUEPYC 7F72AMD EPYC 7F72AMD 7F72160320480640800Min: 923.47 / Avg: 925.98 / Max: 927.91Min: 922.24 / Avg: 929.89 / Max: 944.27Min: 930.22 / Avg: 930.71 / Max: 931.221. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncAMD EPYC 7F72AMD 7F72EPYC 7F7230060090012001500SE +/- 8.92, N = 3SE +/- 4.35, N = 3SE +/- 2.23, N = 31162.711163.461168.581. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncAMD EPYC 7F72AMD 7F72EPYC 7F722004006008001000Min: 1144.88 / Avg: 1162.71 / Max: 1171.66Min: 1154.97 / Avg: 1163.46 / Max: 1169.36Min: 1164.76 / Avg: 1168.58 / Max: 1172.51. (CXX) g++ options: -O3 -lsnappy -lpthread

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F7230060090012001500SE +/- 6.62, N = 3SE +/- 6.46, N = 3SE +/- 7.42, N = 31605.901610.951613.91MIN: 1559.69MIN: 1556.69MIN: 1557.211. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F7230060090012001500Min: 1593.37 / Avg: 1605.9 / Max: 1615.88Min: 1598.24 / Avg: 1610.95 / Max: 1619.3Min: 1603.73 / Avg: 1613.91 / Max: 1628.341. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 100 - Mode: Read OnlyAMD 7F72EPYC 7F72AMD EPYC 7F72150K300K450K600K750KSE +/- 755.05, N = 3SE +/- 1961.28, N = 3SE +/- 1176.24, N = 36987786984526953441. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 100 - Mode: Read OnlyAMD 7F72EPYC 7F72AMD EPYC 7F72120K240K360K480K600KMin: 697404.24 / Avg: 698778.46 / Max: 700007.75Min: 695385.66 / Avg: 698452.18 / Max: 702103.74Min: 692993.28 / Avg: 695344.46 / Max: 696587.511. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-HPLAMD 7F72EPYC 7F72AMD EPYC 7F7220406080100SE +/- 0.30, N = 3SE +/- 0.39, N = 3SE +/- 0.42, N = 387.3187.2586.881. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-HPLAMD 7F72EPYC 7F72AMD EPYC 7F7220406080100Min: 86.73 / Avg: 87.31 / Max: 87.73Min: 86.63 / Avg: 87.25 / Max: 87.98Min: 86.05 / Avg: 86.88 / Max: 87.421. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To CompileEPYC 7F72AMD EPYC 7F72AMD 7F72918273645SE +/- 0.47, N = 3SE +/- 0.37, N = 6SE +/- 0.41, N = 538.8638.9039.05
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To CompileEPYC 7F72AMD EPYC 7F72AMD 7F72816243240Min: 38.08 / Avg: 38.86 / Max: 39.71Min: 37.86 / Avg: 38.9 / Max: 40.63Min: 38.46 / Avg: 39.05 / Max: 40.66

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: SqueezeNetAMD 7F72EPYC 7F72AMD EPYC 7F7220K40K60K80K100KSE +/- 45.73, N = 3SE +/- 44.82, N = 3SE +/- 87.10, N = 389393.689588.189812.7
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetAMD 7F72EPYC 7F72AMD EPYC 7F7216K32K48K64K80KMin: 89313.9 / Avg: 89393.57 / Max: 89472.3Min: 89511.6 / Avg: 89588.07 / Max: 89666.8Min: 89694.8 / Avg: 89812.7 / Max: 89982.7

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 2EPYC 7F72AMD EPYC 7F72AMD 7F7248121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 316.9016.9316.981. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 2EPYC 7F72AMD EPYC 7F72AMD 7F7248121620Min: 16.88 / Avg: 16.9 / Max: 16.93Min: 16.9 / Avg: 16.93 / Max: 16.94Min: 16.95 / Avg: 16.98 / Max: 171. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: MediumEPYC 7F72AMD EPYC 7F72AMD 7F72918273645SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 337.1437.0836.971. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: MediumEPYC 7F72AMD EPYC 7F72AMD 7F72816243240Min: 37.03 / Avg: 37.14 / Max: 37.29Min: 37.04 / Avg: 37.08 / Max: 37.11Min: 36.95 / Avg: 36.97 / Max: 371. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -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.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read OnlyAMD EPYC 7F72AMD 7F72EPYC 7F725K10K15K20K25KSE +/- 325.52, N = 3SE +/- 144.26, N = 3SE +/- 171.06, N = 32444624345243361. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read OnlyAMD EPYC 7F72AMD 7F72EPYC 7F724K8K12K16K20KMin: 23836.86 / Avg: 24446.39 / Max: 24949.24Min: 24131.41 / Avg: 24345.41 / Max: 24620.01Min: 24083.48 / Avg: 24336.14 / Max: 24662.241. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

BYTE Unix Benchmark

This is a test of BYTE. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgLPS, More Is BetterBYTE Unix Benchmark 3.6Computational Test: Dhrystone 2AMD EPYC 7F72AMD 7F72EPYC 7F728M16M24M32M40MSE +/- 93255.90, N = 3SE +/- 231062.98, N = 3SE +/- 295946.74, N = 1237624546.937547272.937455257.2
OpenBenchmarking.orgLPS, More Is BetterBYTE Unix Benchmark 3.6Computational Test: Dhrystone 2AMD EPYC 7F72AMD 7F72EPYC 7F727M14M21M28M35MMin: 37523480.1 / Avg: 37624546.93 / Max: 37810834.2Min: 37101722.6 / Avg: 37547272.9 / Max: 37876274.2Min: 36026099.6 / Avg: 37455257.24 / Max: 39383837.8

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: 1 - Clients: 50 - Mode: Read Write - Average LatencyAMD 7F72AMD EPYC 7F72EPYC 7F72510152025SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 320.5720.6320.661. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average LatencyAMD 7F72AMD EPYC 7F72EPYC 7F72510152025Min: 20.56 / Avg: 20.57 / Max: 20.59Min: 20.56 / Avg: 20.63 / Max: 20.71Min: 20.63 / Avg: 20.66 / Max: 20.711. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: ETC1SEPYC 7F72AMD 7F72AMD EPYC 7F721122334455SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 349.3649.5149.581. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: ETC1SEPYC 7F72AMD 7F72AMD EPYC 7F721020304050Min: 49.34 / Avg: 49.36 / Max: 49.38Min: 49.47 / Avg: 49.51 / Max: 49.54Min: 49.51 / Avg: 49.58 / Max: 49.71. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionAMD 7F72EPYC 7F72AMD EPYC 7F72918273645SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 339.1739.2839.341. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest CompressionAMD 7F72EPYC 7F72AMD EPYC 7F72816243240Min: 39.09 / Avg: 39.17 / Max: 39.33Min: 39.1 / Avg: 39.28 / Max: 39.38Min: 39.17 / Avg: 39.34 / Max: 39.491. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteAMD EPYC 7F72AMD 7F72EPYC 7F72612182430SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 323.223.223.11. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteAMD EPYC 7F72AMD 7F72EPYC 7F72510152025Min: 23.1 / Avg: 23.2 / Max: 23.3Min: 23.1 / Avg: 23.2 / Max: 23.3Min: 23.1 / Avg: 23.13 / Max: 23.21. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillEPYC 7F72AMD EPYC 7F72AMD 7F72612182430SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 323.323.223.21. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillEPYC 7F72AMD EPYC 7F72AMD 7F72510152025Min: 23.2 / Avg: 23.27 / Max: 23.3Min: 23.2 / Avg: 23.2 / Max: 23.2Min: 23.2 / Avg: 23.2 / Max: 23.21. (CXX) g++ options: -O3 -lsnappy -lpthread

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 0.4 AlphaSpeed: 10EPYC 7F72AMD 7F72AMD EPYC 7F720.68381.36762.05142.73523.419SE +/- 0.003, N = 3SE +/- 0.008, N = 3SE +/- 0.002, N = 33.0393.0323.026
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 10EPYC 7F72AMD 7F72AMD EPYC 7F72246810Min: 3.03 / Avg: 3.04 / Max: 3.04Min: 3.02 / Avg: 3.03 / Max: 3.05Min: 3.02 / Avg: 3.03 / Max: 3.03

LibRaw

LibRaw is a RAW image decoder for digital camera photos. This test profile runs LibRaw's post-processing benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing BenchmarkEPYC 7F72AMD 7F72AMD EPYC 7F72816243240SE +/- 0.11, N = 3SE +/- 0.10, N = 3SE +/- 0.07, N = 335.1135.0134.961. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm
OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing BenchmarkEPYC 7F72AMD 7F72AMD EPYC 7F72816243240Min: 34.99 / Avg: 35.11 / Max: 35.32Min: 34.83 / Avg: 35.01 / Max: 35.18Min: 34.84 / Avg: 34.96 / Max: 35.071. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra FastAMD EPYC 7F72AMD 7F72EPYC 7F721020304050SE +/- 0.16, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 342.1742.1341.991. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra FastAMD EPYC 7F72AMD 7F72EPYC 7F72918273645Min: 41.86 / Avg: 42.17 / Max: 42.37Min: 41.98 / Avg: 42.13 / Max: 42.22Min: 41.85 / Avg: 41.99 / Max: 42.061. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100AMD 7F72AMD EPYC 7F72EPYC 7F720.58411.16821.75232.33642.9205SE +/- 0.003, N = 3SE +/- 0.003, N = 3SE +/- 0.005, N = 32.5852.5912.5961. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100AMD 7F72AMD EPYC 7F72EPYC 7F72246810Min: 2.58 / Avg: 2.58 / Max: 2.59Min: 2.58 / Avg: 2.59 / Max: 2.6Min: 2.59 / Avg: 2.6 / Max: 2.61. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillEPYC 7F72AMD EPYC 7F72AMD 7F72612182430SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 324.124.024.01. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillEPYC 7F72AMD EPYC 7F72AMD 7F72612182430Min: 24.1 / Avg: 24.1 / Max: 24.1Min: 24 / Avg: 24.03 / Max: 24.1Min: 23.9 / Avg: 24 / Max: 24.11. (CXX) g++ options: -O3 -lsnappy -lpthread

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read WriteAMD 7F72AMD EPYC 7F72EPYC 7F725001000150020002500SE +/- 1.25, N = 3SE +/- 5.02, N = 3SE +/- 2.57, N = 32431242424211. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read WriteAMD 7F72AMD EPYC 7F72EPYC 7F72400800120016002000Min: 2429 / Avg: 2431.4 / Max: 2433.2Min: 2415.02 / Avg: 2424.41 / Max: 2432.16Min: 2415.51 / Avg: 2420.57 / Max: 2423.941. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Compression SpeedAMD EPYC 7F72EPYC 7F72AMD 7F722K4K6K8K10KSE +/- 17.93, N = 3SE +/- 29.36, N = 3SE +/- 22.56, N = 39780.389777.439740.161. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Compression SpeedAMD EPYC 7F72EPYC 7F72AMD 7F722K4K6K8K10KMin: 9751.48 / Avg: 9780.38 / Max: 9813.23Min: 9722.63 / Avg: 9777.43 / Max: 9823.08Min: 9710.44 / Avg: 9740.16 / Max: 9784.421. (CC) gcc options: -O3

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU-v2-v2 - Model: mobilenet-v2EPYC 7F72AMD 7F72AMD EPYC 7F723691215SE +/- 0.17, N = 3SE +/- 0.17, N = 3SE +/- 0.04, N = 310.0410.0610.08MIN: 8.78 / MAX: 67.63MIN: 8.8 / MAX: 95.72MIN: 8.99 / MAX: 20.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU-v2-v2 - Model: mobilenet-v2EPYC 7F72AMD 7F72AMD EPYC 7F723691215Min: 9.73 / Avg: 10.04 / Max: 10.32Min: 9.8 / Avg: 10.06 / Max: 10.39Min: 10.02 / Avg: 10.08 / Max: 10.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 3EPYC 7F72AMD EPYC 7F72AMD 7F72612182430SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 327.1827.2127.291. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 3EPYC 7F72AMD EPYC 7F72AMD 7F72612182430Min: 27.16 / Avg: 27.18 / Max: 27.2Min: 27.17 / Avg: 27.21 / Max: 27.25Min: 27.26 / Avg: 27.29 / Max: 27.321. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Decompression SpeedAMD EPYC 7F72EPYC 7F72AMD 7F722K4K6K8K10KSE +/- 28.90, N = 3SE +/- 35.72, N = 3SE +/- 3.46, N = 311314.611308.911271.41. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Decompression SpeedAMD EPYC 7F72EPYC 7F72AMD 7F722K4K6K8K10KMin: 11258.5 / Avg: 11314.6 / Max: 11354.7Min: 11273.1 / Avg: 11308.87 / Max: 11380.3Min: 11266.3 / Avg: 11271.4 / Max: 112781. (CC) gcc options: -O3

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: FLOATAMD EPYC 7F72AMD 7F72EPYC 7F7270M140M210M280M350MSE +/- 256408.73, N = 3SE +/- 81095.29, N = 3SE +/- 739610.19, N = 3322432634.13321880722.64321206770.061. (CC) gcc options: -O3 -march=native -lm
OpenBenchmarking.orgQUIPs, More Is BetterHierarchical INTegration 1.0Test: FLOATAMD EPYC 7F72AMD 7F72EPYC 7F7260M120M180M240M300MMin: 322139793.26 / Avg: 322432634.13 / Max: 322943635.8Min: 321790838.8 / Avg: 321880722.64 / Max: 322042583.37Min: 319737613.96 / Avg: 321206770.06 / Max: 322090529.131. (CC) gcc options: -O3 -march=native -lm

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteEPYC 7F72AMD EPYC 7F72AMD 7F7250100150200250SE +/- 0.57, N = 3SE +/- 0.20, N = 3SE +/- 0.41, N = 3209.85210.33210.651. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteEPYC 7F72AMD EPYC 7F72AMD 7F724080120160200Min: 208.72 / Avg: 209.85 / Max: 210.59Min: 210.04 / Avg: 210.33 / Max: 210.71Min: 210.14 / Avg: 210.65 / Max: 211.451. (CXX) g++ options: -O3 -lsnappy -lpthread

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F720.70891.41782.12672.83563.5445SE +/- 0.01523, N = 3SE +/- 0.01746, N = 3SE +/- 0.01172, N = 33.138783.144993.15047MIN: 2.98MIN: 2.97MIN: 2.981. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUAMD EPYC 7F72AMD 7F72EPYC 7F72246810Min: 3.11 / Avg: 3.14 / Max: 3.17Min: 3.11 / Avg: 3.14 / Max: 3.16Min: 3.13 / Avg: 3.15 / Max: 3.171. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: MediumEPYC 7F72AMD EPYC 7F72AMD 7F723691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 310.9510.9210.911. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: MediumEPYC 7F72AMD EPYC 7F72AMD 7F723691215Min: 10.93 / Avg: 10.95 / Max: 10.96Min: 10.91 / Avg: 10.92 / Max: 10.93Min: 10.91 / Avg: 10.91 / Max: 10.911. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 V2EPYC 7F72AMD EPYC 7F72AMD 7F72300K600K900K1200K1500KSE +/- 2727.68, N = 3SE +/- 3330.08, N = 3SE +/- 2105.81, N = 3117566311785901179913
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2EPYC 7F72AMD EPYC 7F72AMD 7F72200K400K600K800K1000KMin: 1170500 / Avg: 1175663.33 / Max: 1179770Min: 1171930 / Avg: 1178590 / Max: 1181960Min: 1175780 / Avg: 1179913.33 / Max: 1182680

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUAMD EPYC 7F72EPYC 7F72AMD 7F720.39010.78021.17031.56041.9505SE +/- 0.00210, N = 3SE +/- 0.00365, N = 3SE +/- 0.01016, N = 31.727651.730511.73388MIN: 1.58MIN: 1.58MIN: 1.571. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUAMD EPYC 7F72EPYC 7F72AMD 7F72246810Min: 1.73 / Avg: 1.73 / Max: 1.73Min: 1.73 / Avg: 1.73 / Max: 1.74Min: 1.72 / Avg: 1.73 / Max: 1.751. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Ultra FastEPYC 7F72AMD EPYC 7F72AMD 7F72306090120150SE +/- 0.27, N = 3SE +/- 0.12, N = 3SE +/- 0.44, N = 3142.44142.03141.931. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Ultra FastEPYC 7F72AMD EPYC 7F72AMD 7F72306090120150Min: 141.89 / Avg: 142.44 / Max: 142.76Min: 141.86 / Avg: 142.03 / Max: 142.27Min: 141.09 / Avg: 141.93 / Max: 142.581. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 V4EPYC 7F72AMD EPYC 7F72AMD 7F72300K600K900K1200K1500KSE +/- 3318.77, N = 3SE +/- 3249.51, N = 3SE +/- 2838.53, N = 3134287713448931347400
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4EPYC 7F72AMD EPYC 7F72AMD 7F72200K400K600K800K1000KMin: 1338640 / Avg: 1342876.67 / Max: 1349420Min: 1339450 / Avg: 1344893.33 / Max: 1350690Min: 1344320 / Avg: 1347400 / Max: 1353070

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read WriteAMD EPYC 7F72EPYC 7F72AMD 7F72400800120016002000SE +/- 5.85, N = 3SE +/- 1.53, N = 3SE +/- 10.89, N = 31807180518011. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read WriteAMD EPYC 7F72EPYC 7F72AMD 7F7230060090012001500Min: 1800.51 / Avg: 1806.58 / Max: 1818.27Min: 1803.34 / Avg: 1805.1 / Max: 1808.14Min: 1784.12 / Avg: 1800.71 / Max: 1821.231. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: SlowEPYC 7F72AMD EPYC 7F72AMD 7F72816243240SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 336.2136.1536.091. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: SlowEPYC 7F72AMD EPYC 7F72AMD 7F72816243240Min: 36.07 / Avg: 36.21 / Max: 36.29Min: 36.11 / Avg: 36.15 / Max: 36.18Min: 35.99 / Avg: 36.09 / Max: 36.21. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 4KAMD EPYC 7F72AMD 7F72EPYC 7F72612182430SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 323.6023.5923.531. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KAMD EPYC 7F72AMD 7F72EPYC 7F72612182430Min: 23.56 / Avg: 23.6 / Max: 23.65Min: 23.49 / Avg: 23.59 / Max: 23.67Min: 23.49 / Avg: 23.53 / Max: 23.591. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Very FastEPYC 7F72AMD 7F72AMD EPYC 7F72612182430SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 324.0023.9823.931. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Very FastEPYC 7F72AMD 7F72AMD EPYC 7F72612182430Min: 23.97 / Avg: 24 / Max: 24.04Min: 23.93 / Avg: 23.98 / Max: 24.01Min: 23.88 / Avg: 23.93 / Max: 23.981. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

IndigoBench

This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: SupercarAMD 7F72AMD EPYC 7F72EPYC 7F723691215SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 310.3410.3110.31
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: SupercarAMD 7F72AMD EPYC 7F72EPYC 7F723691215Min: 10.31 / Avg: 10.34 / Max: 10.38Min: 10.28 / Avg: 10.31 / Max: 10.36Min: 10.23 / Avg: 10.31 / Max: 10.42

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read OnlyAMD 7F72EPYC 7F72AMD EPYC 7F72110K220K330K440K550KSE +/- 493.86, N = 3SE +/- 602.59, N = 3SE +/- 613.62, N = 35305815293185290661. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read OnlyAMD 7F72EPYC 7F72AMD EPYC 7F7290K180K270K360K450KMin: 529788.36 / Avg: 530581.01 / Max: 531487.71Min: 528306.36 / Avg: 529317.62 / Max: 530391Min: 527842.41 / Avg: 529066.11 / Max: 529758.711. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: SlowAMD EPYC 7F72EPYC 7F72AMD 7F723691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 310.7310.7310.701. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: SlowAMD EPYC 7F72EPYC 7F72AMD 7F723691215Min: 10.72 / Avg: 10.73 / Max: 10.74Min: 10.72 / Avg: 10.73 / Max: 10.74Min: 10.69 / Avg: 10.7 / Max: 10.711. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

eSpeak-NG Speech Engine

This test times how long it takes the eSpeak speech synthesizer to read Project Gutenberg's The Outline of Science and output to a WAV file. This test profile is now tracking the eSpeak-NG version of eSpeak. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech SynthesisAMD EPYC 7F72AMD 7F72EPYC 7F72816243240SE +/- 0.09, N = 4SE +/- 0.11, N = 4SE +/- 0.11, N = 432.7132.7632.801. (CC) gcc options: -O2 -std=c99
OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech SynthesisAMD EPYC 7F72AMD 7F72EPYC 7F72714212835Min: 32.45 / Avg: 32.71 / Max: 32.86Min: 32.49 / Avg: 32.76 / Max: 32.97Min: 32.5 / Avg: 32.79 / Max: 33.051. (CC) gcc options: -O2 -std=c99

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillEPYC 7F72AMD EPYC 7F72AMD 7F7250100150200250SE +/- 0.24, N = 3SE +/- 0.12, N = 3SE +/- 0.18, N = 3228.04228.32228.611. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillEPYC 7F72AMD EPYC 7F72AMD 7F724080120160200Min: 227.7 / Avg: 228.04 / Max: 228.51Min: 228.13 / Avg: 228.32 / Max: 228.54Min: 228.28 / Avg: 228.61 / Max: 228.911. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillEPYC 7F72AMD 7F72AMD EPYC 7F7250100150200250SE +/- 0.18, N = 3SE +/- 0.58, N = 3SE +/- 0.16, N = 3220.33220.80220.871. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillEPYC 7F72AMD 7F72AMD EPYC 7F724080120160200Min: 219.99 / Avg: 220.33 / Max: 220.6Min: 219.82 / Avg: 220.8 / Max: 221.82Min: 220.55 / Avg: 220.87 / Max: 221.041. (CXX) g++ options: -O3 -lsnappy -lpthread

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read OnlyAMD EPYC 7F72EPYC 7F72AMD 7F726K12K18K24K30KSE +/- 335.87, N = 4SE +/- 161.30, N = 3SE +/- 325.35, N = 32946229419293911. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read OnlyAMD EPYC 7F72EPYC 7F72AMD 7F725K10K15K20K25KMin: 28740.66 / Avg: 29461.77 / Max: 30350.34Min: 29111.99 / Avg: 29419.1 / Max: 29658.21Min: 29003.71 / Avg: 29391.47 / Max: 30037.891. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

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 AtomsAMD 7F72EPYC 7F72AMD EPYC 7F7248121620SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 315.8215.7915.781. (CXX) g++ options: -O3 -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsAMD 7F72EPYC 7F72AMD EPYC 7F7248121620Min: 15.71 / Avg: 15.82 / Max: 15.96Min: 15.75 / Avg: 15.79 / Max: 15.86Min: 15.76 / Avg: 15.78 / Max: 15.81. (CXX) g++ options: -O3 -pthread -lm

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read WriteAMD EPYC 7F72AMD 7F72EPYC 7F726K12K18K24K30KSE +/- 34.89, N = 3SE +/- 108.48, N = 3SE +/- 54.62, N = 32955229537294861. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read WriteAMD EPYC 7F72AMD 7F72EPYC 7F725K10K15K20K25KMin: 29511.54 / Avg: 29552.27 / Max: 29621.7Min: 29338.64 / Avg: 29536.83 / Max: 29712.38Min: 29406.99 / Avg: 29486.49 / Max: 29591.131. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average LatencyAMD 7F72AMD EPYC 7F72EPYC 7F720.5241.0481.5722.0962.62SE +/- 0.001, N = 3SE +/- 0.003, N = 3SE +/- 0.003, N = 32.3242.3282.3291. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average LatencyAMD 7F72AMD EPYC 7F72EPYC 7F72246810Min: 2.32 / Avg: 2.32 / Max: 2.33Min: 2.32 / Avg: 2.33 / Max: 2.34Min: 2.33 / Avg: 2.33 / Max: 2.331. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average LatencyAMD EPYC 7F72AMD 7F72EPYC 7F720.76391.52782.29173.05563.8195SE +/- 0.003, N = 3SE +/- 0.013, N = 3SE +/- 0.006, N = 33.3883.3893.3951. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average LatencyAMD EPYC 7F72AMD 7F72EPYC 7F72246810Min: 3.38 / Avg: 3.39 / Max: 3.39Min: 3.37 / Avg: 3.39 / Max: 3.41Min: 3.38 / Avg: 3.4 / Max: 3.41. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

InfluxDB

This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000AMD 7F72EPYC 7F72AMD EPYC 7F72300K600K900K1200K1500KSE +/- 3879.14, N = 3SE +/- 2428.05, N = 3SE +/- 2015.92, N = 31199592.41197999.81197198.0
OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000AMD 7F72EPYC 7F72AMD EPYC 7F72200K400K600K800K1000KMin: 1192621.9 / Avg: 1199592.4 / Max: 1206027.6Min: 1194895.1 / Avg: 1197999.83 / Max: 1202785.9Min: 1194830.9 / Avg: 1197198 / Max: 1201208.1

RNNoise

RNNoise is a recurrent neural network for audio noise reduction developed by Mozilla and Xiph.Org. This test profile is a single-threaded test measuring the time to denoise a sample 26 minute long 16-bit RAW audio file using this recurrent neural network noise suppression library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28AMD 7F72EPYC 7F72AMD EPYC 7F72510152025SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 321.1321.1421.171. (CC) gcc options: -O2 -pedantic -fvisibility=hidden
OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28AMD 7F72EPYC 7F72AMD EPYC 7F72510152025Min: 21.12 / Avg: 21.13 / Max: 21.14Min: 21.13 / Avg: 21.14 / Max: 21.15Min: 21.15 / Avg: 21.17 / Max: 21.21. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read WriteAMD 7F72AMD EPYC 7F72EPYC 7F725K10K15K20K25KSE +/- 9.19, N = 3SE +/- 31.97, N = 3SE +/- 24.14, N = 32152121481214791. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read WriteAMD 7F72AMD EPYC 7F72EPYC 7F724K8K12K16K20KMin: 21503.14 / Avg: 21521.45 / Max: 21532.05Min: 21418.59 / Avg: 21481.09 / Max: 21523.99Min: 21430.62 / Avg: 21478.69 / Max: 21506.641. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icaEPYC 7F72AMD 7F72AMD EPYC 7F721224364860SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 351.7451.8251.84
OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icaEPYC 7F72AMD 7F72AMD EPYC 7F721020304050Min: 51.53 / Avg: 51.74 / Max: 51.88Min: 51.67 / Avg: 51.82 / Max: 51.96Min: 51.81 / Avg: 51.84 / Max: 51.88

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteAMD 7F72AMD EPYC 7F72EPYC 7F7250100150200250SE +/- 0.71, N = 3SE +/- 0.57, N = 3SE +/- 0.36, N = 3228.70228.71229.121. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteAMD 7F72AMD EPYC 7F72EPYC 7F724080120160200Min: 227.46 / Avg: 228.7 / Max: 229.9Min: 227.57 / Avg: 228.71 / Max: 229.35Min: 228.43 / Avg: 229.12 / Max: 229.661. (CXX) g++ options: -O3 -lsnappy -lpthread

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 2 + RDO Post-ProcessingAMD 7F72EPYC 7F72AMD EPYC 7F72150300450600750SE +/- 0.27, N = 3SE +/- 2.14, N = 3SE +/- 1.41, N = 3693.90694.87695.001. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 2 + RDO Post-ProcessingAMD 7F72EPYC 7F72AMD EPYC 7F72120240360480600Min: 693.4 / Avg: 693.9 / Max: 694.35Min: 692.38 / Avg: 694.87 / Max: 699.13Min: 692.18 / Avg: 695 / Max: 696.411. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUEPYC 7F72AMD EPYC 7F72AMD 7F7230060090012001500SE +/- 2.75, N = 3SE +/- 3.90, N = 3SE +/- 10.75, N = 31611.211611.741613.70MIN: 1572.74MIN: 1572.94MIN: 1565.851. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUEPYC 7F72AMD EPYC 7F72AMD 7F7230060090012001500Min: 1606.72 / Avg: 1611.21 / Max: 1616.2Min: 1607.18 / Avg: 1611.74 / Max: 1619.5Min: 1593.97 / Avg: 1613.7 / Max: 1630.961. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Timed Clash Compilation

Build the clash-lang Haskell to VHDL/Verilog/SystemVerilog compiler with GHC 8.10.1 Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Clash CompilationTime To CompileAMD EPYC 7F72EPYC 7F72AMD 7F72100200300400500SE +/- 0.52, N = 3SE +/- 0.12, N = 3SE +/- 1.06, N = 3461.62462.28462.32
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Clash CompilationTime To CompileAMD EPYC 7F72EPYC 7F72AMD 7F7280160240320400Min: 460.95 / Avg: 461.62 / Max: 462.64Min: 462.16 / Avg: 462.28 / Max: 462.51Min: 460.54 / Avg: 462.32 / Max: 464.19

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessAMD EPYC 7F72AMD 7F72EPYC 7F72510152025SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 319.0219.0219.041. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessAMD EPYC 7F72AMD 7F72EPYC 7F72510152025Min: 19.01 / Avg: 19.02 / Max: 19.03Min: 19 / Avg: 19.02 / Max: 19.05Min: 19.02 / Avg: 19.04 / Max: 19.081. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 100 - Mode: Read WriteEPYC 7F72AMD 7F72AMD EPYC 7F725001000150020002500SE +/- 4.79, N = 3SE +/- 6.61, N = 3SE +/- 3.93, N = 32145214421421. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 100 - Mode: Read WriteEPYC 7F72AMD 7F72AMD EPYC 7F72400800120016002000Min: 2139.05 / Avg: 2144.57 / Max: 2154.12Min: 2137.32 / Avg: 2144.44 / Max: 2157.66Min: 2135.33 / Avg: 2141.66 / Max: 2148.861. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionEPYC 7F72AMD EPYC 7F72AMD 7F72246810SE +/- 0.004, N = 3SE +/- 0.001, N = 3SE +/- 0.021, N = 38.5388.5488.5491. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest CompressionEPYC 7F72AMD EPYC 7F72AMD 7F723691215Min: 8.53 / Avg: 8.54 / Max: 8.55Min: 8.55 / Avg: 8.55 / Max: 8.55Min: 8.53 / Avg: 8.55 / Max: 8.591. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

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: 1 - Clients: 100 - Mode: Read Write - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F721122334455SE +/- 0.11, N = 3SE +/- 0.14, N = 3SE +/- 0.08, N = 346.6846.6946.741. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 100 - Mode: Read Write - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F721020304050Min: 46.47 / Avg: 46.68 / Max: 46.81Min: 46.42 / Avg: 46.69 / Max: 46.85Min: 46.59 / Avg: 46.74 / Max: 46.871. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

IndigoBench

This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: BedroomAMD 7F72AMD EPYC 7F72EPYC 7F721.07442.14883.22324.29765.372SE +/- 0.013, N = 3SE +/- 0.006, N = 3SE +/- 0.029, N = 34.7754.7704.769
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: CPU - Scene: BedroomAMD 7F72AMD EPYC 7F72EPYC 7F72246810Min: 4.75 / Avg: 4.78 / Max: 4.79Min: 4.76 / Avg: 4.77 / Max: 4.78Min: 4.74 / Avg: 4.77 / Max: 4.83

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1AMD EPYC 7F72EPYC 7F72AMD 7F7260120180240300SE +/- 0.38, N = 3SE +/- 0.13, N = 3SE +/- 0.24, N = 3275.38275.52275.70MIN: 273.95 / MAX: 287.65MIN: 274.32 / MAX: 281.55MIN: 274.17 / MAX: 289.541. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1AMD EPYC 7F72EPYC 7F72AMD 7F7250100150200250Min: 274.62 / Avg: 275.38 / Max: 275.79Min: 275.39 / Avg: 275.52 / Max: 275.78Min: 275.36 / Avg: 275.7 / Max: 276.171. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl

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 0.4 AlphaSpeed: 5AMD EPYC 7F72EPYC 7F72AMD 7F720.23330.46660.69990.93321.1665SE +/- 0.002, N = 3SE +/- 0.003, N = 3SE +/- 0.001, N = 31.0371.0371.036
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 5AMD EPYC 7F72EPYC 7F72AMD 7F72246810Min: 1.03 / Avg: 1.04 / Max: 1.04Min: 1.03 / Avg: 1.04 / Max: 1.04Min: 1.04 / Avg: 1.04 / Max: 1.04

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 6AMD 7F72AMD EPYC 7F72EPYC 7F720.31190.62380.93571.24761.5595SE +/- 0.002, N = 3SE +/- 0.003, N = 3SE +/- 0.004, N = 31.3861.3851.385
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 6AMD 7F72AMD EPYC 7F72EPYC 7F72246810Min: 1.38 / Avg: 1.39 / Max: 1.39Min: 1.38 / Avg: 1.39 / Max: 1.39Min: 1.38 / Avg: 1.38 / Max: 1.39

Timed HMMer Search

This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchAMD EPYC 7F72EPYC 7F72AMD 7F72306090120150SE +/- 0.14, N = 3SE +/- 0.11, N = 3SE +/- 0.02, N = 3142.08142.12142.131. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchAMD EPYC 7F72EPYC 7F72AMD 7F72306090120150Min: 141.82 / Avg: 142.08 / Max: 142.3Min: 141.96 / Avg: 142.11 / Max: 142.33Min: 142.09 / Avg: 142.13 / Max: 142.151. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

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: 100 - Clients: 100 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F720.04250.0850.12750.170.2125SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1890.1890.1891. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F7212345Min: 0.19 / Avg: 0.19 / Max: 0.19Min: 0.19 / Avg: 0.19 / Max: 0.19Min: 0.19 / Avg: 0.19 / Max: 0.191. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F720.00920.01840.02760.03680.046SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.001, N = 30.0410.0410.0411. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F7212345Min: 0.04 / Avg: 0.04 / Max: 0.04Min: 0.04 / Avg: 0.04 / Max: 0.04Min: 0.04 / Avg: 0.04 / Max: 0.041. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F720.01330.02660.03990.05320.0665SE +/- 0.001, N = 3SE +/- 0.000, N = 3SE +/- 0.001, N = 30.0590.0590.0591. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F7212345Min: 0.06 / Avg: 0.06 / Max: 0.06Min: 0.06 / Avg: 0.06 / Max: 0.06Min: 0.06 / Avg: 0.06 / Max: 0.061. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F720.00770.01540.02310.03080.0385SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 40.0340.0340.0341. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average LatencyEPYC 7F72AMD 7F72AMD EPYC 7F7212345Min: 0.03 / Avg: 0.03 / Max: 0.03Min: 0.03 / Avg: 0.03 / Max: 0.03Min: 0.03 / Avg: 0.03 / Max: 0.041. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncAMD EPYC 7F72AMD 7F72EPYC 7F721.01252.0253.03754.055.0625SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 34.54.54.51. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncAMD EPYC 7F72AMD 7F72EPYC 7F72246810Min: 4.5 / Avg: 4.5 / Max: 4.5Min: 4.5 / Avg: 4.5 / Max: 4.5Min: 4.5 / Avg: 4.5 / Max: 4.51. (CXX) g++ options: -O3 -lsnappy -lpthread

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mnasnetAMD EPYC 7F72AMD 7F72EPYC 7F723691215SE +/- 0.18, N = 3SE +/- 0.24, N = 3SE +/- 0.33, N = 39.509.519.58MIN: 8.28 / MAX: 26.52MIN: 8.23 / MAX: 82.21MIN: 8.12 / MAX: 77.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mnasnetAMD EPYC 7F72AMD 7F72EPYC 7F723691215Min: 9.31 / Avg: 9.5 / Max: 9.87Min: 9.09 / Avg: 9.51 / Max: 9.93Min: 9.06 / Avg: 9.58 / Max: 10.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: shufflenet-v2AMD 7F72EPYC 7F72AMD EPYC 7F723691215SE +/- 0.14, N = 3SE +/- 0.49, N = 3SE +/- 0.16, N = 310.3510.4510.45MIN: 8.97 / MAX: 20.34MIN: 9 / MAX: 67.14MIN: 9.13 / MAX: 68.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: shufflenet-v2AMD 7F72EPYC 7F72AMD EPYC 7F723691215Min: 10.1 / Avg: 10.35 / Max: 10.6Min: 9.84 / Avg: 10.45 / Max: 11.41Min: 10.14 / Avg: 10.45 / Max: 10.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Redis

Redis is an open-source data structure server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SETAMD 7F72EPYC 7F72AMD EPYC 7F72300K600K900K1200K1500KSE +/- 35684.24, N = 15SE +/- 37419.69, N = 12SE +/- 27970.44, N = 151500640.311495615.591475849.681. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SETAMD 7F72EPYC 7F72AMD EPYC 7F72300K600K900K1200K1500KMin: 1351394.62 / Avg: 1500640.31 / Max: 1706539.12Min: 1358739.12 / Avg: 1495615.59 / Max: 1727171Min: 1293661 / Avg: 1475849.68 / Max: 1686340.621. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: GETEPYC 7F72AMD EPYC 7F72AMD 7F72400K800K1200K1600K2000KSE +/- 51555.46, N = 15SE +/- 51919.80, N = 15SE +/- 29329.60, N = 152038392.691897703.881814383.801. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: GETEPYC 7F72AMD EPYC 7F72AMD 7F72400K800K1200K1600K2000KMin: 1789023.25 / Avg: 2038392.69 / Max: 2359849Min: 1637080.25 / Avg: 1897703.88 / Max: 2252252.25Min: 1650270.62 / Avg: 1814383.8 / Max: 2020460.621. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: LPUSHAMD 7F72AMD EPYC 7F72EPYC 7F72300K600K900K1200K1500KSE +/- 24808.96, N = 12SE +/- 23963.24, N = 15SE +/- 23366.84, N = 151332610.561317592.971295998.651. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: LPUSHAMD 7F72AMD EPYC 7F72EPYC 7F72200K400K600K800K1000KMin: 1191933.25 / Avg: 1332610.56 / Max: 1449460.88Min: 1228697.75 / Avg: 1317592.97 / Max: 1491028.25Min: 1176470.5 / Avg: 1295998.65 / Max: 1475020.751. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SADDAMD 7F72AMD EPYC 7F72EPYC 7F72400K800K1200K1600K2000KSE +/- 39190.67, N = 15SE +/- 39087.64, N = 15SE +/- 37861.09, N = 121757206.721723521.291658090.041. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SADDAMD 7F72AMD EPYC 7F72EPYC 7F72300K600K900K1200K1500KMin: 1508295.62 / Avg: 1757206.72 / Max: 2012201.12Min: 1538609.25 / Avg: 1723521.29 / Max: 1961286.25Min: 1490456 / Avg: 1658090.04 / Max: 18728991. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: LPOPEPYC 7F72AMD EPYC 7F72AMD 7F72500K1000K1500K2000K2500KSE +/- 44675.79, N = 15SE +/- 61951.41, N = 13SE +/- 22561.14, N = 152134890.151379285.461376509.231. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: LPOPEPYC 7F72AMD EPYC 7F72AMD 7F72400K800K1200K1600K2000KMin: 1938232.62 / Avg: 2134890.15 / Max: 2475247.5Min: 1213864.12 / Avg: 1379285.46 / Max: 2053716.62Min: 1261034.12 / Avg: 1376509.23 / Max: 1525024.381. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7F72EPYC 7F72AMD 7F720.18880.37760.56640.75520.944SE +/- 0.000920, N = 3SE +/- 0.002974, N = 3SE +/- 0.017578, N = 120.5917020.5928040.839025MIN: 0.5MIN: 0.51MIN: 0.641. (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: u8s8f32 - Engine: CPUAMD EPYC 7F72EPYC 7F72AMD 7F72246810Min: 0.59 / Avg: 0.59 / Max: 0.59Min: 0.59 / Avg: 0.59 / Max: 0.6Min: 0.74 / Avg: 0.84 / Max: 0.931. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Max Ping Pong BandwidthAMD EPYC 7F72AMD 7F72EPYC 7F722K4K6K8K10KSE +/- 1135.29, N = 3SE +/- 760.03, N = 3SE +/- 564.95, N = 310456.2910110.269566.501. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Max Ping Pong BandwidthAMD EPYC 7F72AMD 7F72EPYC 7F722K4K6K8K10KMin: 8498.65 / Avg: 10456.29 / Max: 12431.28Min: 8702.82 / Avg: 10110.26 / Max: 11311.26Min: 8667.35 / Avg: 9566.5 / Max: 10608.641. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: EP-STREAM TriadAMD 7F72AMD EPYC 7F72EPYC 7F720.7611.5222.2833.0443.805SE +/- 0.00767, N = 3SE +/- 0.06266, N = 3SE +/- 0.13922, N = 33.382203.299233.116271. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: EP-STREAM TriadAMD 7F72AMD EPYC 7F72EPYC 7F72246810Min: 3.37 / Avg: 3.38 / Max: 3.39Min: 3.17 / Avg: 3.3 / Max: 3.36Min: 2.94 / Avg: 3.12 / Max: 3.391. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-PtransAMD EPYC 7F72AMD 7F72EPYC 7F72246810SE +/- 0.32995, N = 3SE +/- 0.42362, N = 3SE +/- 0.43146, N = 38.285558.175537.757391. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-PtransAMD EPYC 7F72AMD 7F72EPYC 7F723691215Min: 7.63 / Avg: 8.29 / Max: 8.63Min: 7.33 / Avg: 8.18 / Max: 8.61Min: 7.29 / Avg: 7.76 / Max: 8.621. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-FfteEPYC 7F72AMD EPYC 7F72AMD 7F723691215SE +/- 1.15527, N = 3SE +/- 0.73072, N = 3SE +/- 0.56601, N = 39.872389.418958.431211. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-FfteEPYC 7F72AMD EPYC 7F72AMD 7F723691215Min: 7.56 / Avg: 9.87 / Max: 11.05Min: 8.31 / Avg: 9.42 / Max: 10.8Min: 7.51 / Avg: 8.43 / Max: 9.461. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. ATLAS + Open MPI 4.0.3

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 3.1AMD EPYC 7F72EPYC 7F72AMD 7F7248121620SE +/- 0.17, N = 3SE +/- 0.25, N = 12SE +/- 0.32, N = 1215.4515.0014.971. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1AMD EPYC 7F72EPYC 7F72AMD 7F7248121620Min: 15.1 / Avg: 15.45 / Max: 15.65Min: 13.09 / Avg: 15 / Max: 15.67Min: 12.94 / Avg: 14.97 / Max: 15.761. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

156 Results Shown

yquake2
NCNN:
  CPU-v3-v3 - mobilenet-v3
  CPU - mobilenet
LAMMPS Molecular Dynamics Simulator
NCNN
HPC Challenge
InfluxDB
LevelDB
LZ4 Compression
Timed LLVM Compilation
BRL-CAD
LZ4 Compression
NCNN:
  CPU - vgg16
  CPU - efficientnet-b0
AI Benchmark Alpha
NCNN
oneDNN
Mlpack Benchmark
Basis Universal
NCNN:
  CPU - resnet50
  CPU - resnet18
FFTE
NAMD
Mlpack Benchmark
NCNN
Numpy Benchmark
LevelDB
HPC Challenge
NCNN
AI Benchmark Alpha
Stockfish
Mlpack Benchmark
asmFish
NCNN
oneDNN
LevelDB
PostgreSQL pgbench
oneDNN:
  Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU
  IP Shapes 3D - f32 - CPU
  Convolution Batch Shapes Auto - u8s8f32 - CPU
PostgreSQL pgbench
LZ4 Compression
PostgreSQL pgbench
Kvazaar
PHPBench
HPC Challenge
oneDNN
Hugin
PostgreSQL pgbench
oneDNN
rav1e
TensorFlow Lite
oneDNN
TNN
oneDNN
HPC Challenge
oneDNN
PostgreSQL pgbench:
  1 - 100 - Read Only - Average Latency
  100 - 50 - Read Only
AI Benchmark Alpha
LZ4 Compression
Crafty
WebP Image Encode
TensorFlow Lite
x264
GROMACS
x265
oneDNN
TensorFlow Lite
PostgreSQL pgbench
KeyDB
oneDNN:
  Matrix Multiply Batch Shapes Transformer - f32 - CPU
  Recurrent Neural Network Inference - f32 - CPU
LevelDB
oneDNN
PostgreSQL pgbench
HPC Challenge
Timed Linux Kernel Compilation
TensorFlow Lite
Basis Universal
Kvazaar
PostgreSQL pgbench
BYTE Unix Benchmark
PostgreSQL pgbench
Basis Universal
WebP Image Encode
LevelDB:
  Overwrite
  Rand Fill
rav1e
LibRaw
Kvazaar
WebP Image Encode
LevelDB
PostgreSQL pgbench
LZ4 Compression
NCNN
Basis Universal
LZ4 Compression
Hierarchical INTegration
LevelDB
oneDNN
Kvazaar
TensorFlow Lite
oneDNN
Kvazaar
TensorFlow Lite
PostgreSQL pgbench
Kvazaar
x265
Kvazaar
IndigoBench
PostgreSQL pgbench
Kvazaar
eSpeak-NG Speech Engine
LevelDB:
  Rand Fill
  Seq Fill
PostgreSQL pgbench
LAMMPS Molecular Dynamics Simulator
PostgreSQL pgbench:
  100 - 100 - Read Write
  100 - 50 - Read Write - Average Latency
  100 - 100 - Read Write - Average Latency
InfluxDB
RNNoise
PostgreSQL pgbench
Mlpack Benchmark
LevelDB
Basis Universal
oneDNN
Timed Clash Compilation
WebP Image Encode
PostgreSQL pgbench
WebP Image Encode
PostgreSQL pgbench
IndigoBench
TNN
rav1e:
  5
  6
Timed HMMer Search
PostgreSQL pgbench:
  100 - 100 - Read Only - Average Latency
  100 - 1 - Read Only - Average Latency
  1 - 50 - Read Only - Average Latency
  1 - 1 - Read Only - Average Latency
LevelDB
NCNN:
  CPU - mnasnet
  CPU - shufflenet-v2
Redis:
  SET
  GET
  LPUSH
  SADD
  LPOP
oneDNN
HPC Challenge:
  Max Ping Pong Bandwidth
  EP-STREAM Triad
  G-Ptrans
  G-Ffte
High Performance Conjugate Gradient

EPYC 7F72

Compiler Notes: --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 Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301034
Python Notes: Python 3.8.6
Security Notes: 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

Testing initiated at 10 December 2020 20:36 by user phoronix.

AMD 7F72

Compiler Notes: --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 Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301034
Python Notes: Python 3.8.6
Security Notes: 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

Testing initiated at 11 December 2020 09:08 by user phoronix.

AMD EPYC 7F72

Processor: AMD EPYC 7F72 24-Core @ 3.20GHz (24 Cores / 48 Threads), Motherboard: Supermicro H11DSi-NT v2.00 (2.1 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Western Digital WD_BLACK SN850 1TB, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel 10G X550T

OS: Ubuntu 20.10, Kernel: 5.8.0-29-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.5 Mesa 20.2.1 (LLVM 11.0.0 256 bits), Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080

Compiler Notes: --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 Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301034
Python Notes: Python 3.8.6
Security Notes: 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

Testing initiated at 11 December 2020 21:06 by user phoronix.