AMD EPYC 7702 Linux 5.11

AMD EPYC 7702 64-Core testing with a ASRockRack EPYCD8 (P2.10 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 2012245-HA-AMDEPYC7772
Jump To Table - Results

View

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

Limit displaying results to tests within:

Audio Encoding 3 Tests
Bioinformatics 3 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 6 Tests
C/C++ Compiler Tests 15 Tests
Compression Tests 2 Tests
CPU Massive 23 Tests
Creator Workloads 18 Tests
Database Test Suite 7 Tests
Encoding 5 Tests
Fortran Tests 4 Tests
Game Development 5 Tests
Go Language Tests 2 Tests
HPC - High Performance Computing 17 Tests
Common Kernel Benchmarks 5 Tests
Machine Learning 9 Tests
Molecular Dynamics 3 Tests
MPI Benchmarks 3 Tests
Multi-Core 23 Tests
Networking Test Suite 2 Tests
NVIDIA GPU Compute 5 Tests
Intel oneAPI 5 Tests
OpenMPI Tests 3 Tests
Programmer / Developer System Benchmarks 10 Tests
Python Tests 5 Tests
Raytracing 2 Tests
Renderers 5 Tests
Scientific Computing 7 Tests
Server 11 Tests
Server CPU Tests 13 Tests
Single-Threaded 4 Tests
Texture Compression 2 Tests
Video Encoding 2 Tests
Common Workstation Benchmarks 3 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

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
Linux 5.10.2
December 22 2020
  18 Hours, 11 Minutes
Linux 5.11 Git
December 23 2020
  19 Hours, 32 Minutes
Invert Hiding All Results Option
  18 Hours, 52 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


AMD EPYC 7702 Linux 5.11OpenBenchmarking.orgPhoronix Test SuiteAMD EPYC 7702 64-Core @ 2.00GHz (64 Cores / 128 Threads)ASRockRack EPYCD8 (P2.10 BIOS)AMD Starship/Matisse126GB280GB INTEL SSDPED1D280GAllvmpipeAMD Starship/MatisseVE2282 x Intel I350Ubuntu 20.105.10.2-051002-generic (x86_64)5.10.0-051000daily20201222-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 10.2.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelsDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionAMD EPYC 7702 Linux 5.11 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 schedutil (Boost: Enabled) - CPU Microcode: 0x830101c - 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: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Linux 5.10.2 vs. Linux 5.11 Git ComparisonPhoronix Test SuiteBaseline+13.7%+13.7%+27.4%+27.4%+41.1%+41.1%54.8%54.7%24.8%22.5%14.9%14.9%13.1%13%7.6%7.6%7.5%6.2%6%5.8%5.3%5.2%5.1%4.8%4.6%4.2%4.2%4.2%4.1%4.1%4%3.5%3.5%3.4%3.1%3%2.5%2.1%2.1%2%100 - 100 - Read Write100 - 100 - Read Write - Average LatencyTCP - Connections/s - 6439.9%Memorial33.6%Bosphorus 1080p - Ultra Fast25.7%100 - 50 - Read Write - Average Latency100 - 50 - Read WriteSan Miguel - SciVis21.2%100 - 250 - Read Write100 - 250 - Read Write - Average LatencyRand Write - IO_uring - No - Yes - 4KB14.2%Bosphorus 4K14%NASA Streamlines - SciVis13.9%50013.9%1 - 250 - Read Write1 - 250 - Read Write - Average LatencyHTTP - Bandwidth - 6412.8%GET9.2%HTTP - Bandwidth - 19.2%M.R - SciVis9%C.B.S.A - f32 - CPU1 - 50 - Read Write1 - 50 - Read Write - Average LatencyI.R.V7.4%D.I.S6.6%4 - 10000 - 2,5000,1 - 100006.6%NASNet Mobile6.3%Time To CompileD.B.s - f32 - CPU6.1%Rand Fill64 - 10000 - 2,5000,1 - 100005.9%256TCP - Latency - 645.6%Device AI Score5.6%T.T.F.S.SSeq FillCPU - mobilenet512Mobilenet Float4.4%100 - 250 - Read OnlyRand Fill Sync100 - 250 - Read Only - Average Latency1 - 100 - Read Write - Average Latency1 - 100 - Read WriteCPU - yolov4-tinySET3.9%D.T.S3.9%TCP - Latency - 13.9%Bosphorus 4K - Medium3.8%5201 - 10 Seconds - TCP - 643.6%IP Shapes 1D - f32 - CPUIP Shapes 3D - f32 - CPURand Read - IO_uring - No - Yes - 4KB3.5%M.M.B.S.T - f32 - CPUTotal Time3.3%CPU - resnet50R.N.N.T - f32 - CPUWrite2.7%5201 - 10 Seconds - TCP - 322.5%R.N.N.I - f32 - CPU2.5%Inception V4CPU - alexnet2.2%N.2.3.C.F.R2.2%Mobilenet Quant2.1%Bosphorus 4K - Ultra FastRand Read5201 - 10 Seconds - TCP - 12.1%CPU - googlenetPostgreSQL pgbenchPostgreSQL pgbenchEthrIntel Open Image DenoiseKvazaarPostgreSQL pgbenchPostgreSQL pgbenchOSPrayPostgreSQL pgbenchPostgreSQL pgbenchFlexible IO Testerx265OSPrayApache SiegePostgreSQL pgbenchPostgreSQL pgbenchEthrRedisEthrOSPrayoneDNNPostgreSQL pgbenchPostgreSQL pgbenchTensorFlow LiteAI Benchmark AlphaInfluxDBTensorFlow LiteTimed GDB GNU Debugger CompilationoneDNNFacebook RocksDBInfluxDBMariaDBEthrAI Benchmark AlphaYafaRayFacebook RocksDBNCNNMariaDBNode.js V8 Web Tooling BenchmarkTensorFlow LitePostgreSQL pgbenchFacebook RocksDBPostgreSQL pgbenchPostgreSQL pgbenchPostgreSQL pgbenchNCNNRedisAI Benchmark AlphaEthrKvazaariPerfoneDNNoneDNNFlexible IO TesteroneDNNStockfishNCNNoneDNNBlogBenchiPerfoneDNNTensorFlow LiteNCNNFFTETensorFlow LiteKvazaarFacebook RocksDBiPerfNCNNLinux 5.10.2Linux 5.11 Git

AMD EPYC 7702 Linux 5.11openvkl: vklBenchmarkncnn: CPU - regnety_400mncnn: CPU - squeezenet_ssdncnn: CPU - yolov4-tinyncnn: CPU - resnet50ncnn: CPU - alexnetncnn: CPU - resnet18ncnn: CPU - vgg16ncnn: CPU - googlenetncnn: CPU - blazefacencnn: CPU - efficientnet-b0ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - mobilenetncnn: CPU - mnasnetncnn: CPU - shufflenet-v2ncnn: CPU-v2-v2 - mobilenet-v2mysqlslap: 256mysqlslap: 512blogbench: Readrocksdb: Seq Fillecp-candle: P3B2ai-benchmark: Device AI Scoreai-benchmark: Device Training Scoreai-benchmark: Device Inference Scoreonednn: Recurrent Neural Network Inference - f32 - CPUmysqlslap: 128lczero: BLAStensorflow-lite: NASNet Mobilerocksdb: Rand Readbyte: Dhrystone 2numpy: ecp-candle: P3B1hpcg: hint: FLOATbuild-llvm: Time To Compileincompact3d: Cylinderapache-siege: 500mysqlslap: 64yafaray: Total Time For Sample Scenehmmer: Pfam Database Searchtensorflow-lite: Inception V4keydb: tensorflow-lite: Inception ResNet V2tensorflow-lite: Mobilenet Floatinfluxdb: 4 - 10000 - 2,5000,1 - 10000ospray: San Miguel - SciVisblender: Barbershop - CPU-Onlyospray: XFrog Forest - Path Tracertensorflow-lite: Mobilenet Quantbuild-gdb: Time To Compilepgbench: 1 - 250 - Read Write - Average Latencypgbench: 1 - 250 - Read Writeblender: Pabellon Barcelona - CPU-Onlymrbayes: Primate Phylogeny Analysisblender: Classroom - CPU-Onlynode-web-tooling: pgbench: 100 - 50 - Read Write - Average Latencypgbench: 100 - 50 - Read Writeinfluxdb: 64 - 10000 - 2,5000,1 - 10000ospray: XFrog Forest - SciVisbuild-eigen: Time To Compilebuild-linux-kernel: Time To Compileonednn: Recurrent Neural Network Training - f32 - CPUbuild2: Time To Compilesqlite-speedtest: Timed Time - Size 1,000tnn: CPU - MobileNet v2ospray: San Miguel - Path Tracerv-ray: CPUapache-siege: 250compress-lz4: 9 - Decompression Speedcompress-lz4: 9 - Compression Speedkvazaar: Bosphorus 4K - Mediumcompress-lz4: 3 - Decompression Speedcompress-lz4: 3 - Compression Speedindigobench: CPU - Bedroomospray: NASA Streamlines - Path Tracerindigobench: CPU - Supercartensorflow-lite: SqueezeNetrocksdb: Rand Fill Syncrocksdb: Rand Fillrocksdb: Read While Writingsimdjson: LargeRandblender: Fishy Cat - CPU-Onlyapache-siege: 200simdjson: PartialTweetssimdjson: DistinctUserIDredis: SETredis: GETastcenc: Exhaustiveclomp: Static OMP Speeduppgbench: 100 - 250 - Read Only - Average Latencypgbench: 100 - 250 - Read Onlywebp: Quality 100, Lossless, Highest Compressionblender: BMW27 - CPU-Onlystockfish: Total Timex265: Bosphorus 4Kpgbench: 100 - 50 - Read Only - Average Latencypgbench: 100 - 50 - Read Onlypgbench: 100 - 100 - Read Write - Average Latencypgbench: 100 - 100 - Read Writepgbench: 100 - 250 - Read Write - Average Latencypgbench: 100 - 250 - Read Writepgbench: 100 - 100 - Read Only - Average Latencypgbench: 100 - 100 - Read Onlyhimeno: Poisson Pressure Solvernamd: ATPase Simulation - 327,506 Atomspgbench: 1 - 50 - Read Write - Average Latencypgbench: 1 - 50 - Read Writecompress-zstd: 19fio: Rand Read - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Seq Read - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Rand Read - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Seq Read - IO_uring - No - Yes - 2MB - Default Test Directorycompress-lz4: 1 - Decompression Speedcompress-lz4: 1 - Compression Speedfio: Seq Write - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Seq Write - IO_uring - No - Yes - 2MB - Default Test Directorycompress-zstd: 3fio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directorykvazaar: Bosphorus 4K - Very Fastencode-wavpack: WAV To WavPackospray: NASA Streamlines - SciVisethr: TCP - Connections/s - 64ethr: TCP - Connections/s - 1ethr: HTTP - Bandwidth - 64ethr: HTTP - Bandwidth - 1ethr: TCP - Latency - 1ethr: TCP - Bandwidth - 64ethr: TCP - Latency - 64ethr: UDP - Bandwidth - 64pgbench: 1 - 100 - Read Write - Average Latencypgbench: 1 - 100 - Read Writepgbench: 1 - 250 - Read Only - Average Latencypgbench: 1 - 250 - Read Onlypgbench: 1 - 50 - Read Only - Average Latencypgbench: 1 - 50 - Read Onlypgbench: 1 - 100 - Read Only - Average Latencypgbench: 1 - 100 - Read Onlyencode-ape: WAV To APErnnoise: onednn: IP Shapes 1D - f32 - CPUtnn: CPU - SqueezeNet v1.1webp: Quality 100, Losslessonednn: Deconvolution Batch shapes_1d - f32 - CPUbuild-ffmpeg: Time To Compiledolfyn: Computational Fluid Dynamicsbasis: UASTC Level 3onednn: Convolution Batch Shapes Auto - f32 - CPUx265: Bosphorus 1080pkvazaar: Bosphorus 1080p - Mediumencode-opus: WAV To Opus Encodeecp-candle: P1B2kvazaar: Bosphorus 4K - Ultra Fastbasis: UASTC Level 2embree: Pathtracer ISPC - Crowniperf: 5201 - 10 Seconds - TCP - 64iperf: 5201 - 10 Seconds - TCP - 32iperf: 5201 - 10 Seconds - UDP - 64iperf: 5201 - 10 Seconds - UDP - 32iperf: 5201 - 10 Seconds - TCP - 1iperf: 5201 - 10 Seconds - UDP - 1embree: Pathtracer - Crownembree: Pathtracer ISPC - Asian Dragononednn: IP Shapes 3D - f32 - CPUembree: Pathtracer - Asian Dragononednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUastcenc: Thoroughkvazaar: Bosphorus 1080p - Ultra Fastwebp: Quality 100, Highest Compressionoidn: Memorialospray: Magnetic Reconnection - SciVisastcenc: Mediumkvazaar: Bosphorus 1080p - Very Fastastcenc: Fastffte: N=256, 3D Complex FFT Routinewebp: Quality 100onednn: Deconvolution Batch shapes_3d - f32 - CPUospray: Magnetic Reconnection - Path Tracerwebp: Defaultblogbench: WriteLinux 5.10.2Linux 5.11 Git407.61119.1334.8044.0944.9014.4520.5850.9337.118.6627.1720.4939.6445045813039013161371169.60626989841714953.0175402511100573.219348091034094627.0297.53888.60217.3190289027598.96882240.671228.49175042229.8696067.372197.513898756368951.5472746035842.8615400.155.56163.775.9237873.7132.333117.6562154123.73116.217110.957.561.259404731053828.510.99101.42230.1272272.4679.32779.153358.9214.925898337215.6710174.744.859.2410236.946.418.69316.6718.64361162.320512230577979791210.3357.8536679.620.570.591327784.571783246.1645.4057.50.34672576443.11941.1110053264017.830.0736852482.500400695.317471840.1347490173815.5518970.4936915.5143225145.0344000260710979.89467.7721388147.436300021.4316.02476.924249751201910.35540.2827.71162755.1111111127.63100384.22222222236.07527770.24210359700.0608363760.11289752714.39523.3461.64132303.56621.0602.2506420.97220.78219.2112.4919432.8636.979.27544.85642.9913.54347.1709388533611067.133.6385041.0549.176651.10452.2805052.34880.5519396.25192.449.41926.03406.6292.435.70165890.882312142.8552.834522501.77742995410.04119.6234.5542.3843.5514.7720.9750.4436.378.6127.1420.5137.7121.6917.8123.1947648013040833324331180.13725559471608976.763541254410694619754453334088055.1297.39876.82917.3291289511572.00512240.755226.06313137085.8297463.958197.405877016374729.4678146437432.4577350.045.83163.665.9238685.8124.560104.1112437123.88116.582111.607.911.00949579995181.410.99100.33430.1342206.7478.97978.660361.0894.915916037449.1410184.244.018.9010237.646.478.66916.6718.65860929.821382232397379223230.3357.9136097.510.570.581277766.081633435.5645.3958.40.33275658943.07841.729733283415.640.0746749271.616620174.627542220.1327600463852.4147100.4930314.4313469146.533233326121298130310923.69411.68216810818033.7318000124221.1916.03167.553037750501693.68494.9228.78160576.2222222229.18100864.4444444434.65028900.24210378650.0618287210.11190295514.40823.3361.58525303.47020.9382.2896120.95520.80519.0932.3150332.5736.589.27844.56543.9013.50447.1280374903521867.133.6377241.0549.078751.09502.2027852.05930.5337256.25153.049.41919.4936.716.5691.575.70162384.796799552.8503.006802501.77041860OpenBenchmarking.org

OpenVKL

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

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkLinux 5.10.2Linux 5.11 Git90180270360450SE +/- 0.88, N = 3SE +/- 5.07, N = 5407.61410.04MIN: 1 / MAX: 1582MIN: 1 / MAX: 1653

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 20201218Target: CPU - Model: regnety_400mLinux 5.11 GitLinux 5.10.2306090120150SE +/- 0.72, N = 12SE +/- 0.80, N = 12119.62119.13MIN: 111.82 / MAX: 269.8MIN: 110.61 / MAX: 262.541. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssdLinux 5.10.2Linux 5.11 Git816243240SE +/- 0.12, N = 12SE +/- 0.09, N = 1234.8034.55MIN: 33.62 / MAX: 45.89MIN: 33.33 / MAX: 83.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tinyLinux 5.10.2Linux 5.11 Git1020304050SE +/- 0.51, N = 12SE +/- 0.30, N = 1244.0942.38MIN: 40.24 / MAX: 484.5MIN: 39.66 / MAX: 300.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50Linux 5.10.2Linux 5.11 Git1020304050SE +/- 0.53, N = 12SE +/- 0.39, N = 1244.9043.55MIN: 40.6 / MAX: 52.97MIN: 39.63 / MAX: 66.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnetLinux 5.11 GitLinux 5.10.248121620SE +/- 0.27, N = 12SE +/- 0.12, N = 1214.7714.45MIN: 13.31 / MAX: 118.02MIN: 13.51 / MAX: 20.961. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18Linux 5.11 GitLinux 5.10.2510152025SE +/- 0.17, N = 12SE +/- 0.14, N = 1220.9720.58MIN: 19.4 / MAX: 27.16MIN: 19.16 / MAX: 26.631. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg16Linux 5.10.2Linux 5.11 Git1122334455SE +/- 0.31, N = 12SE +/- 0.46, N = 1250.9350.44MIN: 43.23 / MAX: 236.86MIN: 44.49 / MAX: 107.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenetLinux 5.10.2Linux 5.11 Git918273645SE +/- 0.88, N = 12SE +/- 0.81, N = 1237.1136.37MIN: 30.72 / MAX: 54.63MIN: 31.03 / MAX: 49.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazefaceLinux 5.10.2Linux 5.11 Git246810SE +/- 0.22, N = 12SE +/- 0.24, N = 128.668.61MIN: 7.1 / MAX: 18.94MIN: 7.16 / MAX: 143.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b0Linux 5.10.2Linux 5.11 Git612182430SE +/- 0.81, N = 12SE +/- 0.80, N = 1227.1727.14MIN: 20.36 / MAX: 552.93MIN: 20.61 / MAX: 39.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3Linux 5.11 GitLinux 5.10.2510152025SE +/- 0.85, N = 12SE +/- 0.64, N = 1220.5120.49MIN: 15.36 / MAX: 33.79MIN: 15.45 / MAX: 27.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenetLinux 5.10.2Linux 5.11 Git918273645SE +/- 1.06, N = 12SE +/- 0.65, N = 1239.6437.71MIN: 33.54 / MAX: 169.12MIN: 32.98 / MAX: 168.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnetLinux 5.11 Git510152025SE +/- 0.91, N = 1221.69MIN: 15.73 / MAX: 141.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v2Linux 5.11 Git48121620SE +/- 0.52, N = 1217.81MIN: 14.4 / MAX: 28.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v2Linux 5.11 Git612182430SE +/- 1.00, N = 1223.19MIN: 16.62 / MAX: 37.641. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.5.2Clients: 256Linux 5.10.2Linux 5.11 Git100200300400500SE +/- 0.31, N = 3SE +/- 0.67, N = 34504761. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lcrypt -lz -lm -lssl -lcrypto -ldl

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.5.2Clients: 512Linux 5.10.2Linux 5.11 Git100200300400500SE +/- 0.67, N = 3SE +/- 0.66, N = 34584801. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lcrypt -lz -lm -lssl -lcrypto -ldl

BlogBench

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

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: ReadLinux 5.10.2Linux 5.11 Git300K600K900K1200K1500KSE +/- 6044.88, N = 3SE +/- 10744.31, N = 3130390113040831. (CC) gcc options: -O2 -pthread

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillLinux 5.10.2Linux 5.11 Git70K140K210K280K350KSE +/- 258.95, N = 3SE +/- 392.15, N = 33161373324331. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

ECP-CANDLE

The CANDLE benchmark codes implement deep learning architectures relevant to problems in cancer. These architectures address problems at different biological scales, specifically problems at the molecular, cellular and population scales. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterECP-CANDLE 0.3Benchmark: P3B2Linux 5.11 GitLinux 5.10.2300600900120015001180.141169.61

AI Benchmark Alpha

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

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device AI ScoreLinux 5.11 GitLinux 5.10.2600120018002400300025552698

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device Training ScoreLinux 5.11 GitLinux 5.10.22004006008001000947984

OpenBenchmarking.orgScore, More Is BetterAI Benchmark Alpha 0.1.2Device Inference ScoreLinux 5.11 GitLinux 5.10.240080012001600200016081714

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.22004006008001000SE +/- 11.39, N = 15SE +/- 11.69, N = 15976.76953.02MIN: 903.96MIN: 870.071. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.5.2Clients: 128Linux 5.10.2Linux 5.11 Git120240360480600SE +/- 0.97, N = 3SE +/- 1.36, N = 35405411. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lcrypt -lz -lm -lssl -lcrypto -ldl

LeelaChessZero

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

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: BLASLinux 5.10.2Linux 5.11 Git5001000150020002500SE +/- 16.19, N = 3SE +/- 24.66, N = 3251125441. (CXX) g++ options: -flto -pthread

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileLinux 5.11 GitLinux 5.10.220K40K60K80K100KSE +/- 1189.23, N = 15SE +/- 945.46, N = 15106946.0100573.2

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random ReadLinux 5.10.2Linux 5.11 Git40M80M120M160M200MSE +/- 1934294.32, N = 15SE +/- 1745298.00, N = 151934809101975445331. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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 2Linux 5.11 GitLinux 5.10.27M14M21M28M35MSE +/- 346086.49, N = 3SE +/- 335450.15, N = 1234088055.134094627.0

Numpy Benchmark

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

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkLinux 5.11 GitLinux 5.10.260120180240300SE +/- 1.47, N = 3SE +/- 1.47, N = 3297.39297.53

ECP-CANDLE

The CANDLE benchmark codes implement deep learning architectures relevant to problems in cancer. These architectures address problems at different biological scales, specifically problems at the molecular, cellular and population scales. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterECP-CANDLE 0.3Benchmark: P3B1Linux 5.10.2Linux 5.11 Git2004006008001000888.60876.83

High Performance Conjugate Gradient

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

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Linux 5.10.2Linux 5.11 Git48121620SE +/- 0.00, N = 3SE +/- 0.00, N = 317.3217.331. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

Hierarchical INTegration

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

OpenBenchmarking.orgQUIPs, More Is BetterHierarchical INTegration 1.0Test: FLOATLinux 5.10.2Linux 5.11 Git60M120M180M240M300MSE +/- 3132957.23, N = 3SE +/- 2953119.40, N = 3289027598.97289511572.011. (CC) gcc options: -O3 -march=native -lm

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileLinux 5.11 GitLinux 5.10.250100150200250SE +/- 1.77, N = 3SE +/- 0.86, N = 3240.76240.67

Incompact3D

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

OpenBenchmarking.orgSeconds, Fewer Is BetterIncompact3D 2020-09-17Input: CylinderLinux 5.10.2Linux 5.11 Git50100150200250SE +/- 0.78, N = 3SE +/- 0.79, N = 3228.49226.061. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent -levent_pthreads -lutil -lm -lrt -lz

Apache Siege

This is a test of the Apache web server performance being facilitated by the Siege web server benchmark program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Second, More Is BetterApache Siege 2.4.29Concurrent Users: 500Linux 5.11 GitLinux 5.10.29K18K27K36K45KSE +/- 491.70, N = 3SE +/- 510.52, N = 1537085.8242229.861. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.5.2Clients: 64Linux 5.10.2Linux 5.11 Git2004006008001000SE +/- 0.92, N = 3SE +/- 1.38, N = 39609741. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lcrypt -lz -lm -lssl -lcrypto -ldl

YafaRay

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

OpenBenchmarking.orgSeconds, Fewer Is BetterYafaRay 3.4.1Total Time For Sample SceneLinux 5.10.2Linux 5.11 Git1530456075SE +/- 0.66, N = 15SE +/- 0.25, N = 367.3763.961. (CXX) g++ options: -std=c++11 -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype -lpthread

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 SearchLinux 5.10.2Linux 5.11 Git4080120160200SE +/- 0.15, N = 3SE +/- 0.09, N = 3197.51197.411. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4Linux 5.10.2Linux 5.11 Git200K400K600K800K1000KSE +/- 12094.61, N = 4SE +/- 19304.06, N = 15898756877016

KeyDB

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

OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16Linux 5.10.2Linux 5.11 Git80K160K240K320K400KSE +/- 3030.37, N = 13SE +/- 3936.16, N = 3368951.54374729.461. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

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 V2Linux 5.11 GitLinux 5.10.2200K400K600K800K1000KSE +/- 8742.74, N = 15SE +/- 8809.12, N = 3781464727460

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatLinux 5.11 GitLinux 5.10.28K16K24K32K40KSE +/- 441.55, N = 15SE +/- 372.43, N = 337432.435842.8

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: 10000Linux 5.11 GitLinux 5.10.2130K260K390K520K650KSE +/- 62.87, N = 3SE +/- 697.15, N = 3577350.0615400.1

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisLinux 5.11 GitLinux 5.10.21224364860SE +/- 0.59, N = 15SE +/- 0.00, N = 1545.8355.56MIN: 37.04 / MAX: 55.56MIN: 52.63 / MAX: 62.5

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Barbershop - Compute: CPU-OnlyLinux 5.10.2Linux 5.11 Git4080120160200SE +/- 0.16, N = 3SE +/- 0.01, N = 3163.77163.66

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerLinux 5.10.2Linux 5.11 Git1.3322.6643.9965.3286.66SE +/- 0.00, N = 12SE +/- 0.00, N = 125.925.92MIN: 5.78 / MAX: 5.99MIN: 5.81 / MAX: 5.99

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantLinux 5.11 GitLinux 5.10.28K16K24K32K40KSE +/- 465.80, N = 12SE +/- 511.03, N = 338685.837873.7

Timed GDB GNU Debugger Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 9.1Time To CompileLinux 5.10.2Linux 5.11 Git306090120150SE +/- 0.37, N = 3SE +/- 0.21, N = 3132.33124.56

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: 250 - Mode: Read Write - Average LatencyLinux 5.10.2Linux 5.11 Git306090120150SE +/- 3.07, N = 15SE +/- 3.09, N = 15117.66104.111. (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: 250 - Mode: Read WriteLinux 5.10.2Linux 5.11 Git5001000150020002500SE +/- 58.48, N = 15SE +/- 65.50, N = 15215424371. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Pabellon Barcelona - Compute: CPU-OnlyLinux 5.11 GitLinux 5.10.2306090120150SE +/- 0.29, N = 3SE +/- 0.22, N = 3123.88123.73

Timed MrBayes Analysis

This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisLinux 5.11 GitLinux 5.10.2306090120150SE +/- 0.51, N = 3SE +/- 0.16, N = 3116.58116.221. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Classroom - Compute: CPU-OnlyLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.24, N = 3SE +/- 0.71, N = 3111.60110.95

Node.js V8 Web Tooling Benchmark

Running the V8 project's Web-Tooling-Benchmark under Node.js. The Web-Tooling-Benchmark stresses JavaScript-related workloads common to web developers like Babel and TypeScript and Babylon. This test profile can test the system's JavaScript performance with Node.js. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling BenchmarkLinux 5.10.2Linux 5.11 Git246810SE +/- 0.10, N = 3SE +/- 0.02, N = 37.567.911. Nodejs v12.18.2

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 Write - Average LatencyLinux 5.10.2Linux 5.11 Git0.28330.56660.84991.13321.4165SE +/- 0.045, N = 15SE +/- 0.017, N = 31.2591.0091. (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 WriteLinux 5.10.2Linux 5.11 Git11K22K33K44K55KSE +/- 1525.04, N = 15SE +/- 816.79, N = 340473495791. (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: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.11 GitLinux 5.10.2200K400K600K800K1000KSE +/- 12325.27, N = 3SE +/- 7879.99, N = 3995181.41053828.5

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.00, N = 13SE +/- 0.00, N = 1310.9910.99MIN: 10.87 / MAX: 11.24MIN: 10.64 / MAX: 11.24

Timed Eigen Compilation

This test times how long it takes to build all Eigen examples. The Eigen examples are compiled serially. Eigen is a C++ template library for linear algebra. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To CompileLinux 5.10.2Linux 5.11 Git20406080100SE +/- 0.01, N = 3SE +/- 0.04, N = 3101.42100.33

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 CompileLinux 5.11 GitLinux 5.10.2714212835SE +/- 0.30, N = 9SE +/- 0.28, N = 1030.1330.13

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git5001000150020002500SE +/- 2.54, N = 3SE +/- 2.88, N = 32272.462206.74MIN: 2258.16MIN: 2186.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Build2

This test profile measures the time to bootstrap/install the build2 C++ build toolchain from source. Build2 is a cross-platform build toolchain for C/C++ code and features Cargo-like features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To CompileLinux 5.10.2Linux 5.11 Git20406080100SE +/- 0.12, N = 3SE +/- 0.33, N = 379.3378.98

SQLite Speedtest

This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Linux 5.10.2Linux 5.11 Git20406080100SE +/- 0.02, N = 3SE +/- 0.09, N = 379.1578.661. (CC) gcc options: -O2 -ldl -lz -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 v2Linux 5.11 GitLinux 5.10.280160240320400SE +/- 3.09, N = 15SE +/- 4.89, N = 3361.09358.92MIN: 311.51 / MAX: 519.33MIN: 312.05 / MAX: 450.391. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerLinux 5.11 GitLinux 5.10.21.1072.2143.3214.4285.535SE +/- 0.01, N = 3SE +/- 0.01, N = 34.914.92MIN: 4.88 / MAX: 4.93MIN: 4.88 / MAX: 4.95

Chaos Group V-RAY

This is a test of Chaos Group's V-RAY benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgKsamples, More Is BetterChaos Group V-RAY 4.10.07Mode: CPULinux 5.10.2Linux 5.11 Git13K26K39K52K65KSE +/- 208.68, N = 3SE +/- 92.12, N = 35898359160

Apache Siege

This is a test of the Apache web server performance being facilitated by the Siege web server benchmark program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Second, More Is BetterApache Siege 2.4.29Concurrent Users: 250Linux 5.10.2Linux 5.11 Git8K16K24K32K40KSE +/- 82.66, N = 3SE +/- 254.20, N = 337215.6737449.141. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto

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 SpeedLinux 5.10.2Linux 5.11 Git2K4K6K8K10KSE +/- 25.17, N = 3SE +/- 0.37, N = 310174.710184.21. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedLinux 5.11 GitLinux 5.10.21020304050SE +/- 0.08, N = 3SE +/- 0.74, N = 344.0144.851. (CC) gcc options: -O3

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: MediumLinux 5.11 GitLinux 5.10.23691215SE +/- 0.00, N = 3SE +/- 0.00, N = 38.909.241. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 SpeedLinux 5.10.2Linux 5.11 Git2K4K6K8K10KSE +/- 38.82, N = 3SE +/- 37.09, N = 310236.910237.61. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedLinux 5.10.2Linux 5.11 Git1122334455SE +/- 0.57, N = 3SE +/- 0.35, N = 346.4146.471. (CC) gcc options: -O3

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: BedroomLinux 5.11 GitLinux 5.10.2246810SE +/- 0.020, N = 3SE +/- 0.014, N = 38.6698.693

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.00, N = 12SE +/- 0.00, N = 1216.6716.67MIN: 16.13 / MAX: 16.95MIN: 16.13 / MAX: 16.95

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: SupercarLinux 5.10.2Linux 5.11 Git510152025SE +/- 0.03, N = 3SE +/- 0.02, N = 318.6418.66

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: SqueezeNetLinux 5.10.2Linux 5.11 Git13K26K39K52K65KSE +/- 38.04, N = 3SE +/- 97.63, N = 361162.360929.8

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random Fill SyncLinux 5.10.2Linux 5.11 Git50K100K150K200K250KSE +/- 674.13, N = 3SE +/- 431.46, N = 32051222138221. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillLinux 5.10.2Linux 5.11 Git70K140K210K280K350KSE +/- 544.00, N = 3SE +/- 1165.37, N = 33057793239731. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Read While WritingLinux 5.11 GitLinux 5.10.22M4M6M8M10MSE +/- 41997.92, N = 3SE +/- 37021.72, N = 3792232379791211. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandomLinux 5.10.2Linux 5.11 Git0.07430.14860.22290.29720.3715SE +/- 0.00, N = 3SE +/- 0.00, N = 30.330.331. (CXX) g++ options: -O3 -pthread

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: Fishy Cat - Compute: CPU-OnlyLinux 5.11 GitLinux 5.10.21326395265SE +/- 0.08, N = 3SE +/- 0.22, N = 357.9157.85

Apache Siege

This is a test of the Apache web server performance being facilitated by the Siege web server benchmark program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Second, More Is BetterApache Siege 2.4.29Concurrent Users: 200Linux 5.11 GitLinux 5.10.28K16K24K32K40KSE +/- 229.80, N = 3SE +/- 76.79, N = 336097.5136679.621. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto

simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: PartialTweetsLinux 5.10.2Linux 5.11 Git0.12830.25660.38490.51320.6415SE +/- 0.00, N = 3SE +/- 0.00, N = 30.570.571. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: DistinctUserIDLinux 5.11 GitLinux 5.10.20.13280.26560.39840.53120.664SE +/- 0.00, N = 3SE +/- 0.00, N = 30.580.591. (CXX) g++ options: -O3 -pthread

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: SETLinux 5.11 GitLinux 5.10.2300K600K900K1200K1500KSE +/- 11862.79, N = 15SE +/- 22395.01, N = 151277766.081327784.571. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: GETLinux 5.11 GitLinux 5.10.2400K800K1200K1600K2000KSE +/- 29402.50, N = 15SE +/- 18781.21, N = 151633435.561783246.161. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: ExhaustiveLinux 5.10.2Linux 5.11 Git1020304050SE +/- 0.01, N = 3SE +/- 0.00, N = 345.4045.391. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread

CLOMP

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

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP SpeedupLinux 5.10.2Linux 5.11 Git1326395265SE +/- 0.62, N = 3SE +/- 0.38, N = 357.558.41. (CC) gcc options: -fopenmp -O3 -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: 250 - Mode: Read Only - Average LatencyLinux 5.10.2Linux 5.11 Git0.07790.15580.23370.31160.3895SE +/- 0.003, N = 3SE +/- 0.004, N = 50.3460.3321. (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: 250 - Mode: Read OnlyLinux 5.10.2Linux 5.11 Git160K320K480K640K800KSE +/- 6769.94, N = 3SE +/- 9148.67, N = 57257647565891. (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, Lossless, Highest CompressionLinux 5.10.2Linux 5.11 Git1020304050SE +/- 0.08, N = 3SE +/- 0.10, N = 343.1243.081. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -lpng16 -ljpeg

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.90Blend File: BMW27 - Compute: CPU-OnlyLinux 5.11 GitLinux 5.10.21020304050SE +/- 0.31, N = 3SE +/- 0.05, N = 341.7241.11

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 TimeLinux 5.11 GitLinux 5.10.220M40M60M80M100MSE +/- 993423.39, N = 3SE +/- 1682844.67, N = 3973328341005326401. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver

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 4KLinux 5.11 GitLinux 5.10.248121620SE +/- 0.01, N = 3SE +/- 0.10, N = 315.6417.831. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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 LatencyLinux 5.11 GitLinux 5.10.20.01670.03340.05010.06680.0835SE +/- 0.000, N = 3SE +/- 0.001, N = 30.0740.0731. (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 OnlyLinux 5.11 GitLinux 5.10.2150K300K450K600K750KSE +/- 3054.22, N = 3SE +/- 4175.78, N = 36749276852481. (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 LatencyLinux 5.10.2Linux 5.11 Git0.56251.1251.68752.252.8125SE +/- 0.004, N = 3SE +/- 0.001, N = 32.5001.6161. (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 WriteLinux 5.10.2Linux 5.11 Git13K26K39K52K65KSE +/- 70.13, N = 3SE +/- 55.23, N = 340069620171. (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: 250 - Mode: Read Write - Average LatencyLinux 5.10.2Linux 5.11 Git1.19632.39263.58894.78525.9815SE +/- 0.014, N = 3SE +/- 0.012, N = 35.3174.6271. (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: 250 - Mode: Read WriteLinux 5.10.2Linux 5.11 Git12K24K36K48K60KSE +/- 108.81, N = 3SE +/- 148.55, N = 347184542221. (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 LatencyLinux 5.10.2Linux 5.11 Git0.03020.06040.09060.12080.151SE +/- 0.000, N = 3SE +/- 0.001, N = 30.1340.1321. (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 OnlyLinux 5.10.2Linux 5.11 Git160K320K480K640K800KSE +/- 2889.01, N = 3SE +/- 3069.22, N = 37490177600461. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Himeno Benchmark

The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverLinux 5.10.2Linux 5.11 Git8001600240032004000SE +/- 35.09, N = 3SE +/- 7.72, N = 33815.553852.411. (CC) gcc options: -O3 -mavx2

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 AtomsLinux 5.10.2Linux 5.11 Git0.11110.22220.33330.44440.5555SE +/- 0.00061, N = 3SE +/- 0.00027, N = 30.493690.49303

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 LatencyLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.13, N = 3SE +/- 0.19, N = 515.5114.431. (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 WriteLinux 5.10.2Linux 5.11 Git7001400210028003500SE +/- 27.10, N = 3SE +/- 45.00, N = 5322534691. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Zstd Compression

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

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 19Linux 5.10.2Linux 5.11 Git306090120150SE +/- 0.40, N = 3SE +/- 0.72, N = 3145.0146.51. (CC) gcc options: -O3 -pthread -lz -llzma

Flexible IO Tester

FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.270K140K210K280K350KSE +/- 4255.72, N = 3SE +/- 2886.75, N = 33323333440001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git6001200180024003000SE +/- 1.20, N = 3260726121. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 Git30060090012001500SE +/- 15.93, N = 312981. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 Git3006009001200150013031. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

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 SpeedLinux 5.11 GitLinux 5.10.22K4K6K8K10KSE +/- 33.88, N = 3SE +/- 24.24, N = 310923.610979.81. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Compression SpeedLinux 5.11 GitLinux 5.10.22K4K6K8K10KSE +/- 19.19, N = 3SE +/- 52.21, N = 39411.689467.771. (CC) gcc options: -O3

Flexible IO Tester

FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git5001000150020002500SE +/- 13.09, N = 3SE +/- 4.93, N = 3213821681. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 Git2004006008001000SE +/- 2.60, N = 310811. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

Zstd Compression

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

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 3Linux 5.11 GitLinux 5.10.22K4K6K8K10KSE +/- 27.51, N = 3SE +/- 29.74, N = 38033.78147.41. (CC) gcc options: -O3 -pthread -lz -llzma

Flexible IO Tester

FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.280K160K240K320K400KSE +/- 4041.45, N = 3SE +/- 3605.55, N = 33180003630001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 Git30060090012001500SE +/- 14.85, N = 312421. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

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 FastLinux 5.11 GitLinux 5.10.2510152025SE +/- 0.02, N = 3SE +/- 0.04, N = 321.1921.431. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

WavPack Audio Encoding

This test times how long it takes to encode a sample WAV file to WavPack format with very high quality settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPackLinux 5.11 GitLinux 5.10.248121620SE +/- 0.01, N = 5SE +/- 0.01, N = 516.0316.021. (CXX) g++ options: -rdynamic

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVisLinux 5.11 GitLinux 5.10.220406080100SE +/- 2.06, N = 15SE +/- 0.00, N = 1267.5576.92MIN: 50 / MAX: 83.33MIN: 66.67 / MAX: 83.33

Ethr

Ethr is a cross-platform Golang-written network performance measurement tool developed by Microsoft that is capable of testing multiple protocols and different measurements. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 64Linux 5.11 GitLinux 5.10.29K18K27K36K45KSE +/- 200.03, N = 3SE +/- 715.41, N = 33037742497

OpenBenchmarking.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 1Linux 5.11 GitLinux 5.10.211002200330044005500SE +/- 10.00, N = 3SE +/- 32.15, N = 350505120

OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 64Linux 5.11 GitLinux 5.10.2400800120016002000SE +/- 4.48, N = 3SE +/- 3.87, N = 31693.681910.35MIN: 1540 / MAX: 1800MIN: 1780 / MAX: 2000

OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1Linux 5.11 GitLinux 5.10.2120240360480600SE +/- 6.82, N = 3SE +/- 1.90, N = 3494.92540.28MIN: 443.78 / MAX: 537.86MIN: 510.46 / MAX: 553.47

OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1Linux 5.11 GitLinux 5.10.2714212835SE +/- 0.04, N = 3SE +/- 0.07, N = 328.7827.71MIN: 23.95 / MAX: 35.69MIN: 10.95 / MAX: 37.28

OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 64Linux 5.11 GitLinux 5.10.230K60K90K120K150KSE +/- 65.27, N = 3SE +/- 71.49, N = 3160576.22162755.11MIN: 158660 / MAX: 161990MIN: 160640 / MAX: 163900

OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 64Linux 5.11 GitLinux 5.10.2714212835SE +/- 0.06, N = 3SE +/- 0.35, N = 329.1827.63MIN: 24.4 / MAX: 36.26MIN: 10.86 / MAX: 35.5

OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 64Linux 5.10.2Linux 5.11 Git20K40K60K80K100KSE +/- 589.89, N = 3SE +/- 186.19, N = 3100384.22100864.44MIN: 98660 / MAX: 101420MIN: 100080 / MAX: 101500

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 LatencyLinux 5.10.2Linux 5.11 Git816243240SE +/- 0.11, N = 3SE +/- 0.19, N = 336.0834.651. (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 WriteLinux 5.10.2Linux 5.11 Git6001200180024003000SE +/- 7.98, N = 3SE +/- 15.73, N = 3277728901. (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: 250 - Mode: Read Only - Average LatencyLinux 5.11 GitLinux 5.10.20.05450.1090.16350.2180.2725SE +/- 0.001, N = 3SE +/- 0.001, N = 30.2420.2421. (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: 250 - Mode: Read OnlyLinux 5.10.2Linux 5.11 Git200K400K600K800K1000KSE +/- 4214.29, N = 3SE +/- 4352.37, N = 3103597010378651. (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 LatencyLinux 5.11 GitLinux 5.10.20.01370.02740.04110.05480.0685SE +/- 0.000, N = 3SE +/- 0.000, N = 30.0610.0601. (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 OnlyLinux 5.11 GitLinux 5.10.2200K400K600K800K1000KSE +/- 3697.38, N = 3SE +/- 2292.52, N = 38287218363761. (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 LatencyLinux 5.10.2Linux 5.11 Git0.02520.05040.07560.10080.126SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1120.1111. (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 OnlyLinux 5.10.2Linux 5.11 Git200K400K600K800K1000KSE +/- 1328.47, N = 3SE +/- 2133.88, N = 38975279029551. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Monkey Audio Encoding

This test times how long it takes to encode a sample WAV file to Monkey's Audio APE format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMonkey Audio Encoding 3.99.6WAV To APELinux 5.11 GitLinux 5.10.248121620SE +/- 0.02, N = 5SE +/- 0.02, N = 514.4114.401. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt

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 2020-06-28Linux 5.10.2Linux 5.11 Git612182430SE +/- 0.02, N = 3SE +/- 0.01, N = 323.3523.341. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git0.36930.73861.10791.47721.8465SE +/- 0.02008, N = 6SE +/- 0.00489, N = 31.641321.58525MIN: 1.48MIN: 1.51. (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: SqueezeNet v1.1Linux 5.10.2Linux 5.11 Git70140210280350SE +/- 0.06, N = 3SE +/- 0.07, N = 3303.57303.47MIN: 302.78 / MAX: 304.41MIN: 302.92 / MAX: 304.021. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl

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, LosslessLinux 5.10.2Linux 5.11 Git510152025SE +/- 0.14, N = 3SE +/- 0.01, N = 321.0620.941. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -lpng16 -ljpeg

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.20.51521.03041.54562.06082.576SE +/- 0.03688, N = 3SE +/- 0.03762, N = 32.289612.25064MIN: 2.06MIN: 2.021. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Timed FFmpeg Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.2.2Time To CompileLinux 5.10.2Linux 5.11 Git510152025SE +/- 0.01, N = 3SE +/- 0.01, N = 320.9720.96

Dolfyn

Dolfyn is a Computational Fluid Dynamics (CFD) code of modern numerical simulation techniques. The Dolfyn test profile measures the execution time of the bundled computational fluid dynamics demos that are bundled with Dolfyn. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDolfyn 0.527Computational Fluid DynamicsLinux 5.11 GitLinux 5.10.2510152025SE +/- 0.05, N = 3SE +/- 0.00, N = 320.8120.78

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 3Linux 5.10.2Linux 5.11 Git510152025SE +/- 0.01, N = 3SE +/- 0.01, N = 319.2119.091. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git0.56071.12141.68212.24282.8035SE +/- 0.02807, N = 3SE +/- 0.02725, N = 152.491942.31503MIN: 2.38MIN: 2.11. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 1080pLinux 5.11 GitLinux 5.10.2816243240SE +/- 0.02, N = 3SE +/- 0.05, N = 332.5732.861. (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 1080p - Video Preset: MediumLinux 5.11 GitLinux 5.10.2918273645SE +/- 0.12, N = 3SE +/- 0.06, N = 336.5836.971. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Opus Codec Encoding

Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodeLinux 5.11 GitLinux 5.10.23691215SE +/- 0.014, N = 5SE +/- 0.015, N = 59.2789.2751. (CXX) g++ options: -fvisibility=hidden -logg -lm

ECP-CANDLE

The CANDLE benchmark codes implement deep learning architectures relevant to problems in cancer. These architectures address problems at different biological scales, specifically problems at the molecular, cellular and population scales. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterECP-CANDLE 0.3Benchmark: P1B2Linux 5.10.2Linux 5.11 Git102030405044.8644.57

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 FastLinux 5.10.2Linux 5.11 Git1020304050SE +/- 0.08, N = 3SE +/- 0.09, N = 342.9943.901. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 2Linux 5.10.2Linux 5.11 Git3691215SE +/- 0.02, N = 3SE +/- 0.01, N = 313.5413.501. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: CrownLinux 5.11 GitLinux 5.10.21122334455SE +/- 0.07, N = 3SE +/- 0.14, N = 347.1347.17MIN: 46.12 / MAX: 48.75MIN: 46.21 / MAX: 48.92

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 64Linux 5.11 GitLinux 5.10.28K16K24K32K40KSE +/- 125.92, N = 3SE +/- 326.95, N = 337490388531. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 32Linux 5.11 GitLinux 5.10.28K16K24K32K40KSE +/- 9.54, N = 3SE +/- 526.87, N = 335218361101. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 64Linux 5.10.2Linux 5.11 Git1530456075SE +/- 0.00, N = 3SE +/- 0.00, N = 367.167.11. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 32Linux 5.10.2Linux 5.11 Git816243240SE +/- 0.00, N = 3SE +/- 0.00, N = 333.633.61. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 1Linux 5.11 GitLinux 5.10.28K16K24K32K40KSE +/- 318.15, N = 3SE +/- 170.09, N = 337724385041. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 1Linux 5.10.2Linux 5.11 Git0.23630.47260.70890.94521.1815SE +/- 0.00, N = 3SE +/- 0.00, N = 31.051.051. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: CrownLinux 5.11 GitLinux 5.10.21122334455SE +/- 0.04, N = 3SE +/- 0.12, N = 349.0849.18MIN: 48.12 / MAX: 50.9MIN: 48.06 / MAX: 51.15

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: Asian DragonLinux 5.11 GitLinux 5.10.21224364860SE +/- 0.09, N = 3SE +/- 0.25, N = 351.1051.10MIN: 50.17 / MAX: 53.55MIN: 49.85 / MAX: 53.17

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git0.51311.02621.53932.05242.5655SE +/- 0.00516, N = 3SE +/- 0.02757, N = 52.280502.20278MIN: 2.19MIN: 2.051. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: Asian DragonLinux 5.11 GitLinux 5.10.21224364860SE +/- 0.29, N = 3SE +/- 0.07, N = 352.0652.35MIN: 50.92 / MAX: 53.8MIN: 51.56 / MAX: 54.22

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git0.12420.24840.37260.49680.621SE +/- 0.007913, N = 3SE +/- 0.004029, N = 30.5519390.533725MIN: 0.51MIN: 0.511. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: ThoroughLinux 5.11 GitLinux 5.10.2246810SE +/- 0.00, N = 3SE +/- 0.00, N = 36.256.251. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -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 FastLinux 5.11 GitLinux 5.10.24080120160200SE +/- 6.43, N = 15SE +/- 0.56, N = 3153.04192.441. (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 100, Highest CompressionLinux 5.11 GitLinux 5.10.23691215SE +/- 0.009, N = 3SE +/- 0.006, N = 39.4199.4191. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -lpng16 -ljpeg

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.2.0Scene: MemorialLinux 5.11 GitLinux 5.10.2612182430SE +/- 0.00, N = 3SE +/- 0.02, N = 319.4926.03

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: SciVisLinux 5.11 GitLinux 5.10.2918273645SE +/- 0.33, N = 436.7140.00MIN: 31.25 / MAX: 40MIN: 38.46

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: MediumLinux 5.10.2Linux 5.11 Git246810SE +/- 0.01, N = 3SE +/- 0.01, N = 36.626.561. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -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: Very FastLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.07, N = 3SE +/- 0.53, N = 391.5792.431. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

ASTC Encoder

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

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.0Preset: FastLinux 5.11 GitLinux 5.10.21.28252.5653.84755.136.4125SE +/- 0.01, N = 3SE +/- 0.01, N = 35.705.701. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -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 RoutineLinux 5.11 GitLinux 5.10.240K80K120K160K200KSE +/- 2518.16, N = 3SE +/- 2418.06, N = 3162384.80165890.881. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp

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 100Linux 5.10.2Linux 5.11 Git0.64241.28481.92722.56963.212SE +/- 0.003, N = 3SE +/- 0.001, N = 32.8552.8501. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -lpng16 -ljpeg

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.20.67651.3532.02952.7063.3825SE +/- 0.01141, N = 3SE +/- 0.02719, N = 33.006802.83452MIN: 2.69MIN: 2.681. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: Path TracerLinux 5.10.2Linux 5.11 Git50100150200250250250MIN: 166.67 / MAX: 333.33MIN: 200 / MAX: 333.33

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: DefaultLinux 5.10.2Linux 5.11 Git0.39980.79961.19941.59921.999SE +/- 0.002, N = 3SE +/- 0.004, N = 31.7771.7701. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -lpng16 -ljpeg

BlogBench

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

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: WriteLinux 5.11 GitLinux 5.10.29K18K27K36K45KSE +/- 198.23, N = 3SE +/- 84.40, N = 341860429951. (CC) gcc options: -O2 -pthread

181 Results Shown

OpenVKL
NCNN:
  CPU - regnety_400m
  CPU - squeezenet_ssd
  CPU - yolov4-tiny
  CPU - resnet50
  CPU - alexnet
  CPU - resnet18
  CPU - vgg16
  CPU - googlenet
  CPU - blazeface
  CPU - efficientnet-b0
  CPU-v3-v3 - mobilenet-v3
  CPU - mobilenet
  CPU - mnasnet
  CPU - shufflenet-v2
  CPU-v2-v2 - mobilenet-v2
MariaDB:
  256
  512
BlogBench
Facebook RocksDB
ECP-CANDLE
AI Benchmark Alpha:
  Device AI Score
  Device Training Score
  Device Inference Score
oneDNN
MariaDB
LeelaChessZero
TensorFlow Lite
Facebook RocksDB
BYTE Unix Benchmark
Numpy Benchmark
ECP-CANDLE
High Performance Conjugate Gradient
Hierarchical INTegration
Timed LLVM Compilation
Incompact3D
Apache Siege
MariaDB
YafaRay
Timed HMMer Search
TensorFlow Lite
KeyDB
TensorFlow Lite:
  Inception ResNet V2
  Mobilenet Float
InfluxDB
OSPray
Blender
OSPray
TensorFlow Lite
Timed GDB GNU Debugger Compilation
PostgreSQL pgbench:
  1 - 250 - Read Write - Average Latency
  1 - 250 - Read Write
Blender
Timed MrBayes Analysis
Blender
Node.js V8 Web Tooling Benchmark
PostgreSQL pgbench:
  100 - 50 - Read Write - Average Latency
  100 - 50 - Read Write
InfluxDB
OSPray
Timed Eigen Compilation
Timed Linux Kernel Compilation
oneDNN
Build2
SQLite Speedtest
TNN
OSPray
Chaos Group V-RAY
Apache Siege
LZ4 Compression:
  9 - Decompression Speed
  9 - Compression Speed
Kvazaar
LZ4 Compression:
  3 - Decompression Speed
  3 - Compression Speed
IndigoBench
OSPray
IndigoBench
TensorFlow Lite
Facebook RocksDB:
  Rand Fill Sync
  Rand Fill
  Read While Writing
simdjson
Blender
Apache Siege
simdjson:
  PartialTweets
  DistinctUserID
Redis:
  SET
  GET
ASTC Encoder
CLOMP
PostgreSQL pgbench:
  100 - 250 - Read Only - Average Latency
  100 - 250 - Read Only
WebP Image Encode
Blender
Stockfish
x265
PostgreSQL pgbench:
  100 - 50 - Read Only - Average Latency
  100 - 50 - Read Only
  100 - 100 - Read Write - Average Latency
  100 - 100 - Read Write
  100 - 250 - Read Write - Average Latency
  100 - 250 - Read Write
  100 - 100 - Read Only - Average Latency
  100 - 100 - Read Only
Himeno Benchmark
NAMD
PostgreSQL pgbench:
  1 - 50 - Read Write - Average Latency
  1 - 50 - Read Write
Zstd Compression
Flexible IO Tester:
  Rand Read - IO_uring - No - Yes - 4KB - Default Test Directory
  Seq Read - IO_uring - No - Yes - 2MB - Default Test Directory
  Rand Read - IO_uring - No - Yes - 4KB - Default Test Directory
  Seq Read - IO_uring - No - Yes - 2MB - Default Test Directory
LZ4 Compression:
  1 - Decompression Speed
  1 - Compression Speed
Flexible IO Tester:
  Seq Write - IO_uring - No - Yes - 2MB - Default Test Directory:
    MB/s
    IOPS
Zstd Compression
Flexible IO Tester:
  Rand Write - IO_uring - No - Yes - 4KB - Default Test Directory:
    IOPS
    MB/s
Kvazaar
WavPack Audio Encoding
OSPray
Ethr:
  TCP - Connections/s - 64
  TCP - Connections/s - 1
  HTTP - Bandwidth - 64
  HTTP - Bandwidth - 1
  TCP - Latency - 1
  TCP - Bandwidth - 64
  TCP - Latency - 64
  UDP - Bandwidth - 64
PostgreSQL pgbench:
  1 - 100 - Read Write - Average Latency
  1 - 100 - Read Write
  1 - 250 - Read Only - Average Latency
  1 - 250 - Read Only
  1 - 50 - Read Only - Average Latency
  1 - 50 - Read Only
  1 - 100 - Read Only - Average Latency
  1 - 100 - Read Only
Monkey Audio Encoding
RNNoise
oneDNN
TNN
WebP Image Encode
oneDNN
Timed FFmpeg Compilation
Dolfyn
Basis Universal
oneDNN
x265
Kvazaar
Opus Codec Encoding
ECP-CANDLE
Kvazaar
Basis Universal
Embree
iPerf:
  5201 - 10 Seconds - TCP - 64
  5201 - 10 Seconds - TCP - 32
  5201 - 10 Seconds - UDP - 64
  5201 - 10 Seconds - UDP - 32
  5201 - 10 Seconds - TCP - 1
  5201 - 10 Seconds - UDP - 1
Embree:
  Pathtracer - Crown
  Pathtracer ISPC - Asian Dragon
oneDNN
Embree
oneDNN
ASTC Encoder
Kvazaar
WebP Image Encode
Intel Open Image Denoise
OSPray
ASTC Encoder
Kvazaar
ASTC Encoder
FFTE
WebP Image Encode
oneDNN
OSPray
WebP Image Encode
BlogBench