Xeon E3-1245 v5 Ubuntu Linux 5.4 Intel Xeon E3-1245 v5 testing with a MSI C236A WORKSTATION (MS-7998) v1.0 (2.90 BIOS) and MSI Intel HD P530 3GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2010041-FI-XEONE312446&grs&sor .
Xeon E3-1245 v5 Ubuntu Linux 5.4 Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution Vulkan Linux 5.4 Linux 5.8 Linux 5.9 Git Intel Xeon E3-1245 v5 @ 3.90GHz (4 Cores / 8 Threads) MSI C236A WORKSTATION (MS-7998) v1.0 (2.90 BIOS) Intel Xeon E3-1200 v5/E3-1500 32GB 120GB Samsung SSD 850 MSI Intel HD P530 3GB (1150MHz) Realtek ALC1150 LG Ultra HD Intel I219-V Ubuntu 20.04 5.4.0-37-generic (x86_64) GNOME Shell 3.36.2 X Server 1.20.8 modesetting 1.20.8 4.6 Mesa 20.0.4 GCC 9.3.0 ext4 3840x2160 5.8.13-050813-generic (x86_64) 1.2.131 5.9.0-050900rc7daily20201003-generic (x86_64) 20201002 GNOME Shell 3.36.4 OpenBenchmarking.org Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,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 Processor Details - Scaling Governor: intel_pstate powersave - CPU Microcode: 0xdc Python Details - Python 3.8.2 Security Details - Linux 5.4: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable - Linux 5.8: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable - Linux 5.9 Git: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable
Xeon E3-1245 v5 Ubuntu Linux 5.4 realsr-ncnn: 4x - No ncnn: Vulkan GPU - mobilenet ncnn: Vulkan GPU - squeezenet ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - efficientnet-b0 lczero: BLAS ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPU - alexnet ncnn: Vulkan GPU - mnasnet ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 ncnn: Vulkan GPU - resnet50 glmark2: 3840 x 2160 ncnn: Vulkan GPU - yolov4-tiny couchdb: 100 - 1000 - 24 glmark2: 1920 x 1080 onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Deconvolution Batch deconv_1d - f32 - CPU onednn: Deconvolution Batch deconv_3d - f32 - CPU ncnn: Vulkan GPU - vgg16 mlpack: scikit_ica keydb: onednn: Recurrent Neural Network Training - f32 - CPU crafty: Elapsed Time onednn: IP Batch 1D - u8s8f32 - CPU onednn: IP Batch All - f32 - CPU onednn: IP Batch 1D - f32 - CPU deepspeech: CPU lczero: Eigen onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU onednn: IP Batch All - u8s8f32 - CPU system-decompress-gzip: mlpack: scikit_linearridgeregression build-gdb: Time To Compile onednn: Deconvolution Batch deconv_1d - u8s8f32 - CPU ncnn: CPU - alexnet espeak: Text-To-Speech Synthesis vkfft: montage: Mosaic of M17, K band, 1.5 deg x 1.5 deg hugin: Panorama Photo Assistant + Stitching Time rodinia: OpenMP Leukocyte mnn: resnet-v2-50 avifenc: 8 ncnn: CPU - blazeface ecp-candle: P3B2 mnn: SqueezeNetV1.0 ecp-candle: P3B1 ecp-candle: P1B2 astcenc: Fast webp: Quality 100 avifenc: 10 mnn: inception-v3 byte: Dhrystone 2 gpaw: Carbon Nanotube mlpack: scikit_svm rnnoise: build-apache: Time To Compile tscp: AI Chess Performance avifenc: 2 influxdb: 64 - 10000 - 2,5000,1 - 10000 compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 webp: Quality 100, Lossless ncnn: CPU - mnasnet aom-av1: Speed 8 Realtime tesseract-ocr: Time To OCR 7 Images rodinia: OpenMP LavaMD mnn: mobilenet-v1-1.0 ncnn: CPU - efficientnet-b0 mnn: MobileNetV2_224 build-linux-kernel: Time To Compile tnn: CPU - SqueezeNet v1.1 ncnn: CPU - vgg16 rodinia: OpenMP CFD Solver webp: Quality 100, Lossless, Highest Compression rodinia: OpenMP HotSpot3D mpv: Big Buck Bunny Sunflower 4K - Software Only caffe: GoogleNet - CPU - 200 aom-av1: Speed 4 Two-Pass dolfyn: Computational Fluid Dynamics webp: Default ncnn: CPU-v2-v2 - mobilenet-v2 onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU compress-zstd: 19 ncnn: CPU - squeezenet incompact3d: Cylinder webp: Quality 100, Highest Compression lammps: Rhodopsin Protein avifenc: 0 onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU libraw: Post-Processing Benchmark gromacs: Water Benchmark ffte: N=256, 3D Complex FFT Routine ncnn: CPU - mobilenet aom-av1: Speed 6 Two-Pass mocassin: Dust 2D tau100.0 namd: ATPase Simulation - 327,506 Atoms caffe: GoogleNet - CPU - 100 mpv: Big Buck Bunny Sunflower 1080p - Software Only ncnn: CPU - yolov4-tiny influxdb: 1024 - 10000 - 2,5000,1 - 10000 onednn: Convolution Batch Shapes Auto - f32 - CPU ai-benchmark: Device Inference Score caffe: AlexNet - CPU - 100 ocrmypdf: Processing 60 Page PDF Document ncnn: CPU - resnet18 build-llvm: Time To Compile ncnn: CPU - googlenet ncnn: CPU - shufflenet-v2 blender: BMW27 - CPU-Only mafft: Multiple Sequence Alignment - LSU RNA hmmer: Pfam Database Search onednn: Deconvolution Batch deconv_3d - u8s8f32 - CPU kripke: ncnn: CPU-v3-v3 - mobilenet-v3 astcenc: Thorough ai-benchmark: Device AI Score compress-zstd: 3 tensorflow-lite: Mobilenet Quant hint: FLOAT blender: Fishy Cat - CPU-Only rodinia: OpenMP Streamcluster ncnn: CPU - resnet50 tnn: CPU - MobileNet v2 astcenc: Medium caffe: AlexNet - CPU - 200 tensorflow-lite: NASNet Mobile tensorflow-lite: SqueezeNet astcenc: Exhaustive tensorflow-lite: Inception ResNet V2 tensorflow-lite: Inception V4 tensorflow-lite: Mobilenet Float ai-benchmark: Device Training Score aom-av1: Speed 6 Realtime aom-av1: Speed 0 Two-Pass influxdb: 4 - 10000 - 2,5000,1 - 10000 mlpack: scikit_qda ncnn: Vulkan GPU - blazeface ncnn: Vulkan GPU - shufflenet-v2 Linux 5.4 Linux 5.8 Linux 5.9 Git 17.241 46.43 52.13 42.16 30.86 797 36.80 16.32 57.94 15.11 14.72 87.11 135 89.34 175.791 543 285.961 9.25837 14.3651 214.21 71.10 393603.00 569.397 7547320 3.77996 111.115 8.17233 78.51331 717 5.15748 49.8609 3.282 4.29 159.604 10.3947 22.56 33.933 1141 89.559 73.463 254.675 56.164 8.199 2.17 590.047 10.255 1212.787 46.128 9.02 2.728 7.600 62.308 38538549.1 598.320 27.86 27.870 32.732 1141277 108.057 990095.1 50.735 19.361 6.66 35.82 27.447 570.414 7.748 10.51 5.601 199.635 344.444 89.35 47.393 47.466 118.303 32.10 328888 1.88 21.425 1.741 7.79 7.16575 20.5 24.88 750.917623 8.148 2.746 182.456 18.7529 27.66 0.504 20328.495177544 28.94 3.00 305 3.87433 164645 139.45 39.48 996459.9 19.9426 664 66942 55.538 23.09 1493.222 23.08 4.67 382.67 12.268 125.430 7.41923 17835730 6.53 72.26 1397 2386.0 348999 381569301.12653 532.49 28.734 46.50 366.297 11.14 133823 366947 499694 578.17 6517420 7204167 339412 733 14.99 0.22 870964.9 88.35 2.63 10.11 283.500 38.59 43.80 35.48 26.02 678 31.34 14 51.30 13 12.69 75.25 119 79.45 157.303 502 270.665 9.73579 15.1450 206.48 68.13 381234.65 548.044 7469298 3.90177 114.236 8.37706 80.60955 701 5.04980 50.9712 3.339 4.24 157.323 10.3074 22.12 34.257 1161 90.697 74.056 256.916 56.934 8.257 2.14 598.284 10.373 1226.039 45.575 9.12 2.742 7.639 62.931 38513841.4 597.093 28.12 28.130 32.712 1134693 108.403 984645.3 50.618 19.333 6.61 35.57 27.639 568.094 7.788 10.44 5.638 199.789 346.500 88.95 47.671 47.260 118.459 32.22 330435 1.87 21.476 1.750 7.82 7.19867 20.4 24.76 754.293904 8.184 2.758 182.788 18.8285 27.65 0.504 20258.596219821 28.85 2.99 304 3.88624 165139 139.88 39.36 993437.4 19.8862 663 67067 55.393 23.11 1490.327 23.03 4.66 382.25 12.291 125.495 7.42274 17866357 6.53 72.32 1396 2389.3 348976 381270686.12418 531.91 28.730 46.45 366.393 11.14 133870 366914 499845 578.30 6518137 7204273 339372 733 14.99 0.22 814462.2 84.76 2.35 10.06 283.462 38.17 43.80 35.47 26.02 779 31.36 13.99 49.78 13 12.68 75.27 137 79.33 161.474 552 260.964 9.82786 14.7150 205.15 70.06 377603.40 558.863 7302256 3.84716 110.716 8.40703 80.04749 719 5.04492 50.2447 3.355 4.33 160.636 10.1850 22.41 34.588 1161 91.033 74.650 258.772 57.044 8.316 2.15 592.267 10.396 1229.33 45.976 9.07 2.756 7.678 62.561 38169338.8 602.790 27.87 28.088 33.008 1131430 108.956 982092.1 51.026 19.214 6.62 35.65 27.567 566.532 7.800 10.45 5.605 200.948 346.140 88.82 47.631 47.190 118.977 32.28 330664 1.88 21.537 1.749 7.78 7.20203 20.4 24.84 754.269389 8.161 2.758 183.225 18.7935 27.55 0.502 20335.309333098 28.84 2.99 304 3.88703 165181 139.75 39.45 994952.9 19.8822 665 66888 55.509 23.05 1493.850 23.07 4.66 381.93 12.291 125.660 7.40944 17848217 6.52 72.37 1398 2389.0 349438 381134357.52245 532.18 28.703 46.46 366.631 11.15 133774 367078 499825 578.22 6518793 7202907 339434 733 14.99 0.22 805704.5 86.12 2.29 11.09 OpenBenchmarking.org
RealSR-NCNN Scale: 4x - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No Linux 5.4 Linux 5.9 Git Linux 5.8 60 120 180 240 300 SE +/- 0.24, N = 4 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 17.24 283.46 283.50
NCNN Target: Vulkan GPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: mobilenet Linux 5.9 Git Linux 5.8 Linux 5.4 11 22 33 44 55 SE +/- 0.24, N = 3 SE +/- 0.25, N = 3 SE +/- 0.43, N = 3 38.17 38.59 46.43 MIN: 35.95 / MAX: 45.56 MIN: 36.59 / MAX: 45.66 MIN: 44.53 / MAX: 65.27 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: squeezenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: squeezenet Linux 5.8 Linux 5.9 Git Linux 5.4 12 24 36 48 60 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 43.80 43.80 52.13 MIN: 43.21 / MAX: 44.65 MIN: 43.39 / MAX: 43.96 MIN: 50.27 / MAX: 53.67 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: googlenet Linux 5.9 Git Linux 5.8 Linux 5.4 10 20 30 40 50 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 35.47 35.48 42.16 MIN: 34.79 / MAX: 35.67 MIN: 34.99 / MAX: 35.72 MIN: 38.96 / MAX: 43.23 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: efficientnet-b0 Linux 5.8 Linux 5.9 Git Linux 5.4 7 14 21 28 35 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 26.02 26.02 30.86 MIN: 25.28 / MAX: 26.55 MIN: 25.31 / MAX: 27.35 MIN: 29.36 / MAX: 35.2 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
LeelaChessZero Backend: BLAS OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: BLAS Linux 5.4 Linux 5.9 Git Linux 5.8 200 400 600 800 1000 SE +/- 0.67, N = 3 SE +/- 2.33, N = 3 SE +/- 9.24, N = 3 797 779 678 1. (CXX) g++ options: -flto -pthread
NCNN Target: Vulkan GPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: resnet18 Linux 5.8 Linux 5.9 Git Linux 5.4 8 16 24 32 40 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 31.34 31.36 36.80 MIN: 30.87 / MAX: 31.48 MIN: 30.73 / MAX: 31.56 MIN: 36.41 / MAX: 37.2 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 Linux 5.9 Git Linux 5.8 Linux 5.4 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 13.99 14.00 16.32 MIN: 13.58 / MAX: 14.57 MIN: 13.24 / MAX: 14.29 MIN: 16.25 / MAX: 16.42 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: alexnet Linux 5.9 Git Linux 5.8 Linux 5.4 13 26 39 52 65 SE +/- 0.88, N = 3 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 49.78 51.30 57.94 MIN: 46.19 / MAX: 53.72 MIN: 48.82 / MAX: 54.66 MIN: 55.09 / MAX: 61.52 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: mnasnet Linux 5.8 Linux 5.9 Git Linux 5.4 4 8 12 16 20 SE +/- 0.01, N = 3 13.00 13.00 15.11 MIN: 12.57 / MAX: 13.06 MIN: 12.53 / MAX: 14.37 MIN: 14.91 / MAX: 15.25 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 Linux 5.9 Git Linux 5.8 Linux 5.4 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 12.68 12.69 14.72 MIN: 12.28 / MAX: 12.98 MIN: 11.98 / MAX: 13.56 MIN: 14.5 / MAX: 15.08 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: resnet50 Linux 5.8 Linux 5.9 Git Linux 5.4 20 40 60 80 100 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 75.25 75.27 87.11 MIN: 74.67 / MAX: 76.95 MIN: 74.66 / MAX: 75.49 MIN: 84.69 / MAX: 89 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
GLmark2 Resolution: 3840 x 2160 OpenBenchmarking.org Score, More Is Better GLmark2 2020.04 Resolution: 3840 x 2160 Linux 5.9 Git Linux 5.4 Linux 5.8 30 60 90 120 150 137 135 119
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: yolov4-tiny Linux 5.9 Git Linux 5.8 Linux 5.4 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 79.33 79.45 89.34 MIN: 78.43 / MAX: 80.24 MIN: 78.55 / MAX: 81.51 MIN: 78.43 / MAX: 101.68 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Apache CouchDB Bulk Size: 100 - Inserts: 1000 - Rounds: 24 OpenBenchmarking.org Seconds, Fewer Is Better Apache CouchDB 3.1.1 Bulk Size: 100 - Inserts: 1000 - Rounds: 24 Linux 5.8 Linux 5.9 Git Linux 5.4 40 80 120 160 200 SE +/- 2.35, N = 4 SE +/- 2.13, N = 5 SE +/- 1.34, N = 3 157.30 161.47 175.79 1. (CXX) g++ options: -std=c++14 -lmozjs-68 -lm -lerl_interface -lei -fPIC -MMD
GLmark2 Resolution: 1920 x 1080 OpenBenchmarking.org Score, More Is Better GLmark2 2020.04 Resolution: 1920 x 1080 Linux 5.9 Git Linux 5.4 Linux 5.8 120 240 360 480 600 552 543 502
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU Linux 5.9 Git Linux 5.8 Linux 5.4 60 120 180 240 300 SE +/- 2.39, N = 15 SE +/- 3.43, N = 3 SE +/- 0.67, N = 3 260.96 270.67 285.96 MIN: 247.57 MIN: 261.84 MIN: 283.43 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU Linux 5.4 Linux 5.8 Linux 5.9 Git 3 6 9 12 15 SE +/- 0.00752, N = 3 SE +/- 0.08955, N = 3 SE +/- 0.10703, N = 3 9.25837 9.73579 9.82786 MIN: 9.19 MIN: 9.52 MIN: 9.53 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU Linux 5.4 Linux 5.9 Git Linux 5.8 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 14.37 14.72 15.15 MIN: 14.12 MIN: 14.45 MIN: 14.79 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: vgg16 Linux 5.9 Git Linux 5.8 Linux 5.4 50 100 150 200 250 SE +/- 0.22, N = 3 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 205.15 206.48 214.21 MIN: 203.28 / MAX: 208.47 MIN: 204.27 / MAX: 209.41 MIN: 212.41 / MAX: 218.75 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Mlpack Benchmark Benchmark: scikit_ica OpenBenchmarking.org Seconds, Fewer Is Better Mlpack Benchmark Benchmark: scikit_ica Linux 5.8 Linux 5.9 Git Linux 5.4 16 32 48 64 80 SE +/- 0.91, N = 3 SE +/- 0.43, N = 3 SE +/- 0.60, N = 3 68.13 70.06 71.10
KeyDB OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.0.16 Linux 5.4 Linux 5.8 Linux 5.9 Git 80K 160K 240K 320K 400K SE +/- 767.36, N = 3 SE +/- 387.75, N = 3 SE +/- 359.00, N = 3 393603.00 381234.65 377603.40 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU Linux 5.8 Linux 5.9 Git Linux 5.4 120 240 360 480 600 SE +/- 0.78, N = 3 SE +/- 3.18, N = 3 SE +/- 0.15, N = 3 548.04 558.86 569.40 MIN: 545.86 MIN: 552.13 MIN: 568.18 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Crafty Elapsed Time OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time Linux 5.4 Linux 5.8 Linux 5.9 Git 1.6M 3.2M 4.8M 6.4M 8M SE +/- 35007.79, N = 3 SE +/- 23889.21, N = 3 SE +/- 21668.21, N = 3 7547320 7469298 7302256 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
oneDNN Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU Linux 5.4 Linux 5.9 Git Linux 5.8 0.8779 1.7558 2.6337 3.5116 4.3895 SE +/- 0.00729, N = 3 SE +/- 0.00590, N = 3 SE +/- 0.00186, N = 3 3.77996 3.84716 3.90177 MIN: 3.67 MIN: 3.73 MIN: 3.79 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch All - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: f32 - Engine: CPU Linux 5.9 Git Linux 5.4 Linux 5.8 30 60 90 120 150 SE +/- 0.10, N = 3 SE +/- 0.29, N = 3 SE +/- 1.51, N = 3 110.72 111.12 114.24 MIN: 109.51 MIN: 109.55 MIN: 110.66 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: f32 - Engine: CPU Linux 5.4 Linux 5.8 Linux 5.9 Git 2 4 6 8 10 SE +/- 0.00538, N = 3 SE +/- 0.03078, N = 3 SE +/- 0.03775, N = 3 8.17233 8.37706 8.40703 MIN: 7.82 MIN: 8 MIN: 7.97 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
DeepSpeech Acceleration: CPU OpenBenchmarking.org Seconds, Fewer Is Better DeepSpeech 0.6 Acceleration: CPU Linux 5.4 Linux 5.9 Git Linux 5.8 20 40 60 80 100 SE +/- 0.05, N = 3 SE +/- 0.53, N = 3 SE +/- 0.51, N = 3 78.51 80.05 80.61
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: Eigen Linux 5.9 Git Linux 5.4 Linux 5.8 160 320 480 640 800 SE +/- 3.28, N = 3 SE +/- 5.46, N = 3 SE +/- 10.51, N = 9 719 717 701 1. (CXX) g++ options: -flto -pthread
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU Linux 5.9 Git Linux 5.8 Linux 5.4 1.1604 2.3208 3.4812 4.6416 5.802 SE +/- 0.00900, N = 3 SE +/- 0.01544, N = 3 SE +/- 0.00807, N = 3 5.04492 5.04980 5.15748 MIN: 4.96 MIN: 4.96 MIN: 5.07 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU Linux 5.4 Linux 5.9 Git Linux 5.8 11 22 33 44 55 SE +/- 0.01, N = 3 SE +/- 0.15, N = 3 SE +/- 0.17, N = 3 49.86 50.24 50.97 MIN: 49.38 MIN: 49.56 MIN: 50.11 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
System GZIP Decompression OpenBenchmarking.org Seconds, Fewer Is Better System GZIP Decompression Linux 5.4 Linux 5.8 Linux 5.9 Git 0.7549 1.5098 2.2647 3.0196 3.7745 SE +/- 0.001, N = 3 SE +/- 0.035, N = 14 SE +/- 0.046, N = 13 3.282 3.339 3.355
Mlpack Benchmark Benchmark: scikit_linearridgeregression OpenBenchmarking.org Seconds, Fewer Is Better Mlpack Benchmark Benchmark: scikit_linearridgeregression Linux 5.8 Linux 5.4 Linux 5.9 Git 0.9743 1.9486 2.9229 3.8972 4.8715 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 4 4.24 4.29 4.33
Timed GDB GNU Debugger Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GDB GNU Debugger Compilation 9.1 Time To Compile Linux 5.8 Linux 5.4 Linux 5.9 Git 40 80 120 160 200 SE +/- 0.25, N = 3 SE +/- 0.20, N = 3 SE +/- 0.30, N = 3 157.32 159.60 160.64
oneDNN Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU Linux 5.9 Git Linux 5.8 Linux 5.4 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 10.19 10.31 10.39 MIN: 10.07 MIN: 10.1 MIN: 10.15 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: alexnet Linux 5.8 Linux 5.9 Git Linux 5.4 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.28, N = 3 SE +/- 0.22, N = 3 22.12 22.41 22.56 MIN: 22.04 / MAX: 22.28 MIN: 22.05 / MAX: 59.82 MIN: 22.06 / MAX: 39.49 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
eSpeak-NG Speech Engine Text-To-Speech Synthesis OpenBenchmarking.org Seconds, Fewer Is Better eSpeak-NG Speech Engine 20200907 Text-To-Speech Synthesis Linux 5.4 Linux 5.8 Linux 5.9 Git 8 16 24 32 40 SE +/- 0.22, N = 16 SE +/- 0.20, N = 4 SE +/- 0.13, N = 4 33.93 34.26 34.59 1. (CC) gcc options: -O2 -std=c99
VkFFT OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 2020-09-29 Linux 5.9 Git Linux 5.8 Linux 5.4 200 400 600 800 1000 SE +/- 0.67, N = 3 1161 1161 1141
Montage Astronomical Image Mosaic Engine Mosaic of M17, K band, 1.5 deg x 1.5 deg OpenBenchmarking.org Seconds, Fewer Is Better Montage Astronomical Image Mosaic Engine 6.0 Mosaic of M17, K band, 1.5 deg x 1.5 deg Linux 5.4 Linux 5.8 Linux 5.9 Git 20 40 60 80 100 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.18, N = 3 89.56 90.70 91.03 1. (CC) gcc options: -std=gnu99 -lcfitsio -lm -O2
Hugin Panorama Photo Assistant + Stitching Time OpenBenchmarking.org Seconds, Fewer Is Better Hugin Panorama Photo Assistant + Stitching Time Linux 5.4 Linux 5.8 Linux 5.9 Git 20 40 60 80 100 SE +/- 0.48, N = 3 SE +/- 0.25, N = 3 SE +/- 0.42, N = 3 73.46 74.06 74.65
Rodinia Test: OpenMP Leukocyte OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Leukocyte Linux 5.4 Linux 5.8 Linux 5.9 Git 60 120 180 240 300 SE +/- 0.36, N = 3 SE +/- 0.33, N = 3 SE +/- 1.15, N = 3 254.68 256.92 258.77 1. (CXX) g++ options: -O2 -lOpenCL
Mobile Neural Network Model: resnet-v2-50 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: resnet-v2-50 Linux 5.4 Linux 5.8 Linux 5.9 Git 13 26 39 52 65 SE +/- 0.16, N = 3 SE +/- 0.13, N = 3 SE +/- 0.07, N = 3 56.16 56.93 57.04 MIN: 55.4 / MAX: 107.43 MIN: 56.62 / MAX: 80.52 MIN: 56.76 / MAX: 79.83 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
libavif avifenc Encoder Speed: 8 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.7.3 Encoder Speed: 8 Linux 5.4 Linux 5.8 Linux 5.9 Git 2 4 6 8 10 SE +/- 0.013, N = 3 SE +/- 0.022, N = 3 SE +/- 0.007, N = 3 8.199 8.257 8.316 1. (CXX) g++ options: -O3 -fPIC
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: blazeface Linux 5.8 Linux 5.9 Git Linux 5.4 0.4883 0.9766 1.4649 1.9532 2.4415 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 2.14 2.15 2.17 MIN: 2.13 / MAX: 2.24 MIN: 2.13 / MAX: 2.24 MIN: 2.13 / MAX: 2.29 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
ECP-CANDLE Benchmark: P3B2 OpenBenchmarking.org Seconds, Fewer Is Better ECP-CANDLE 0.3 Benchmark: P3B2 Linux 5.4 Linux 5.9 Git Linux 5.8 130 260 390 520 650 590.05 592.27 598.28
Mobile Neural Network Model: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: SqueezeNetV1.0 Linux 5.4 Linux 5.8 Linux 5.9 Git 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 10.26 10.37 10.40 MIN: 10.11 / MAX: 16.05 MIN: 10.27 / MAX: 31.91 MIN: 10.23 / MAX: 33.26 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
ECP-CANDLE Benchmark: P3B1 OpenBenchmarking.org Seconds, Fewer Is Better ECP-CANDLE 0.3 Benchmark: P3B1 Linux 5.4 Linux 5.8 Linux 5.9 Git 300 600 900 1200 1500 1212.79 1226.04 1229.33
ECP-CANDLE Benchmark: P1B2 OpenBenchmarking.org Seconds, Fewer Is Better ECP-CANDLE 0.3 Benchmark: P1B2 Linux 5.8 Linux 5.9 Git Linux 5.4 10 20 30 40 50 45.58 45.98 46.13
ASTC Encoder Preset: Fast OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Fast Linux 5.4 Linux 5.9 Git Linux 5.8 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 9.02 9.07 9.12 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
WebP Image Encode Encode Settings: Quality 100 OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100 Linux 5.4 Linux 5.8 Linux 5.9 Git 0.6201 1.2402 1.8603 2.4804 3.1005 SE +/- 0.001, N = 3 SE +/- 0.003, N = 3 SE +/- 0.005, N = 3 2.728 2.742 2.756 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
libavif avifenc Encoder Speed: 10 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.7.3 Encoder Speed: 10 Linux 5.4 Linux 5.8 Linux 5.9 Git 2 4 6 8 10 SE +/- 0.017, N = 3 SE +/- 0.030, N = 3 SE +/- 0.031, N = 3 7.600 7.639 7.678 1. (CXX) g++ options: -O3 -fPIC
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: inception-v3 Linux 5.4 Linux 5.9 Git Linux 5.8 14 28 42 56 70 SE +/- 0.15, N = 3 SE +/- 0.64, N = 3 SE +/- 0.20, N = 3 62.31 62.56 62.93 MIN: 61.37 / MAX: 84.45 MIN: 60.97 / MAX: 89.35 MIN: 62.1 / MAX: 89.18 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
BYTE Unix Benchmark Computational Test: Dhrystone 2 OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 3.6 Computational Test: Dhrystone 2 Linux 5.4 Linux 5.8 Linux 5.9 Git 8M 16M 24M 32M 40M SE +/- 54862.86, N = 3 SE +/- 6470.73, N = 3 SE +/- 156718.13, N = 3 38538549.1 38513841.4 38169338.8
GPAW Input: Carbon Nanotube OpenBenchmarking.org Seconds, Fewer Is Better GPAW 20.1 Input: Carbon Nanotube Linux 5.8 Linux 5.4 Linux 5.9 Git 130 260 390 520 650 SE +/- 0.45, N = 3 SE +/- 0.74, N = 3 SE +/- 4.33, N = 3 597.09 598.32 602.79 1. (CC) gcc options: -pthread -shared -fwrapv -O2 -lxc -lblas -lmpi
Mlpack Benchmark Benchmark: scikit_svm OpenBenchmarking.org Seconds, Fewer Is Better Mlpack Benchmark Benchmark: scikit_svm Linux 5.4 Linux 5.9 Git Linux 5.8 7 14 21 28 35 SE +/- 0.10, N = 3 SE +/- 0.00, N = 3 SE +/- 0.13, N = 3 27.86 27.87 28.12
RNNoise OpenBenchmarking.org Seconds, Fewer Is Better RNNoise 2020-06-28 Linux 5.4 Linux 5.9 Git Linux 5.8 7 14 21 28 35 SE +/- 0.07, N = 3 SE +/- 0.07, N = 3 SE +/- 0.18, N = 3 27.87 28.09 28.13 1. (CC) gcc options: -O2 -pedantic -fvisibility=hidden
Timed Apache Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Apache Compilation 2.4.41 Time To Compile Linux 5.8 Linux 5.4 Linux 5.9 Git 8 16 24 32 40 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 32.71 32.73 33.01
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance Linux 5.4 Linux 5.8 Linux 5.9 Git 200K 400K 600K 800K 1000K SE +/- 1206.41, N = 5 SE +/- 1191.39, N = 5 SE +/- 1137.79, N = 5 1141277 1134693 1131430 1. (CC) gcc options: -O3 -march=native
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.7.3 Encoder Speed: 2 Linux 5.4 Linux 5.8 Linux 5.9 Git 20 40 60 80 100 SE +/- 0.16, N = 3 SE +/- 0.10, N = 3 SE +/- 0.18, N = 3 108.06 108.40 108.96 1. (CXX) g++ options: -O3 -fPIC
InfluxDB Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 Linux 5.4 Linux 5.8 Linux 5.9 Git 200K 400K 600K 800K 1000K SE +/- 1019.70, N = 3 SE +/- 2768.93, N = 3 SE +/- 4935.24, N = 3 990095.1 984645.3 982092.1
XZ Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 Linux 5.8 Linux 5.4 Linux 5.9 Git 12 24 36 48 60 SE +/- 0.11, N = 3 SE +/- 0.05, N = 3 SE +/- 0.37, N = 3 50.62 50.74 51.03 1. (CC) gcc options: -pthread -fvisibility=hidden -O2
WebP Image Encode Encode Settings: Quality 100, Lossless OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless Linux 5.9 Git Linux 5.8 Linux 5.4 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 19.21 19.33 19.36 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mnasnet Linux 5.8 Linux 5.9 Git Linux 5.4 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 6.61 6.62 6.66 MIN: 6.56 / MAX: 8.52 MIN: 6.57 / MAX: 8.48 MIN: 6.56 / MAX: 20.23 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
AOM AV1 Encoder Mode: Speed 8 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 8 Realtime Linux 5.4 Linux 5.9 Git Linux 5.8 8 16 24 32 40 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 35.82 35.65 35.57 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Tesseract OCR Time To OCR 7 Images OpenBenchmarking.org Seconds, Fewer Is Better Tesseract OCR 4.1.1 Time To OCR 7 Images Linux 5.4 Linux 5.9 Git Linux 5.8 7 14 21 28 35 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 27.45 27.57 27.64
Rodinia Test: OpenMP LavaMD OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP LavaMD Linux 5.9 Git Linux 5.8 Linux 5.4 120 240 360 480 600 SE +/- 0.14, N = 3 SE +/- 1.01, N = 3 SE +/- 2.84, N = 3 566.53 568.09 570.41 1. (CXX) g++ options: -O2 -lOpenCL
Mobile Neural Network Model: mobilenet-v1-1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: mobilenet-v1-1.0 Linux 5.4 Linux 5.8 Linux 5.9 Git 2 4 6 8 10 SE +/- 0.009, N = 3 SE +/- 0.014, N = 3 SE +/- 0.003, N = 3 7.748 7.788 7.800 MIN: 7.67 / MAX: 22.21 MIN: 7.73 / MAX: 12.65 MIN: 7.74 / MAX: 27.87 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: efficientnet-b0 Linux 5.8 Linux 5.9 Git Linux 5.4 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.06, N = 3 10.44 10.45 10.51 MIN: 10.4 / MAX: 10.76 MIN: 10.4 / MAX: 13.27 MIN: 10.41 / MAX: 11 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: MobileNetV2_224 Linux 5.4 Linux 5.9 Git Linux 5.8 1.2686 2.5372 3.8058 5.0744 6.343 SE +/- 0.019, N = 3 SE +/- 0.037, N = 3 SE +/- 0.018, N = 3 5.601 5.605 5.638 MIN: 5.47 / MAX: 9.61 MIN: 5.45 / MAX: 10.27 MIN: 5.53 / MAX: 13.75 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Timed Linux Kernel Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.4 Time To Compile Linux 5.4 Linux 5.8 Linux 5.9 Git 40 80 120 160 200 SE +/- 0.78, N = 3 SE +/- 0.59, N = 3 SE +/- 0.79, N = 3 199.64 199.79 200.95
TNN Target: CPU - Model: SqueezeNet v1.1 OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: SqueezeNet v1.1 Linux 5.4 Linux 5.9 Git Linux 5.8 80 160 240 320 400 SE +/- 0.08, N = 3 SE +/- 0.40, N = 3 SE +/- 0.04, N = 3 344.44 346.14 346.50 MIN: 343.7 / MAX: 346.67 MIN: 344.07 / MAX: 348.21 MIN: 345.83 / MAX: 347.67 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
NCNN Target: CPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: vgg16 Linux 5.9 Git Linux 5.8 Linux 5.4 20 40 60 80 100 SE +/- 0.07, N = 3 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 88.82 88.95 89.35 MIN: 88.53 / MAX: 101.19 MIN: 88.67 / MAX: 137.54 MIN: 89.05 / MAX: 94.32 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Rodinia Test: OpenMP CFD Solver OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP CFD Solver Linux 5.4 Linux 5.9 Git Linux 5.8 11 22 33 44 55 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 47.39 47.63 47.67 1. (CXX) g++ options: -O2 -lOpenCL
WebP Image Encode Encode Settings: Quality 100, Lossless, Highest Compression OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless, Highest Compression Linux 5.9 Git Linux 5.8 Linux 5.4 11 22 33 44 55 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.14, N = 3 47.19 47.26 47.47 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Rodinia Test: OpenMP HotSpot3D OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP HotSpot3D Linux 5.4 Linux 5.8 Linux 5.9 Git 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.30, N = 3 SE +/- 0.30, N = 3 118.30 118.46 118.98 1. (CXX) g++ options: -O2 -lOpenCL
MPV Video Input: Big Buck Bunny Sunflower 4K - Decode: Software Only OpenBenchmarking.org FPS, More Is Better MPV Video Input: Big Buck Bunny Sunflower 4K - Decode: Software Only Linux 5.9 Git Linux 5.8 Linux 5.4 7 14 21 28 35 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 32.28 32.22 32.10 MIN: 31.58 / MAX: 33.49 MIN: 31.25 / MAX: 33.33 MIN: 31.41 / MAX: 32.79 1. mpv 0.32.0
Caffe Model: GoogleNet - Acceleration: CPU - Iterations: 200 OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: GoogleNet - Acceleration: CPU - Iterations: 200 Linux 5.4 Linux 5.8 Linux 5.9 Git 70K 140K 210K 280K 350K SE +/- 211.61, N = 3 SE +/- 96.33, N = 3 SE +/- 315.26, N = 3 328888 330435 330664 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
AOM AV1 Encoder Mode: Speed 4 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 4 Two-Pass Linux 5.9 Git Linux 5.4 Linux 5.8 0.423 0.846 1.269 1.692 2.115 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.88 1.88 1.87 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Dolfyn Computational Fluid Dynamics OpenBenchmarking.org Seconds, Fewer Is Better Dolfyn 0.527 Computational Fluid Dynamics Linux 5.4 Linux 5.8 Linux 5.9 Git 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 21.43 21.48 21.54
WebP Image Encode Encode Settings: Default OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Default Linux 5.4 Linux 5.9 Git Linux 5.8 0.3938 0.7876 1.1814 1.5752 1.969 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 1.741 1.749 1.750 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
NCNN Target: CPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU-v2-v2 - Model: mobilenet-v2 Linux 5.9 Git Linux 5.4 Linux 5.8 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 7.78 7.79 7.82 MIN: 7.67 / MAX: 9.94 MIN: 7.67 / MAX: 9.99 MIN: 7.68 / MAX: 20.52 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU Linux 5.4 Linux 5.8 Linux 5.9 Git 2 4 6 8 10 SE +/- 0.00597, N = 3 SE +/- 0.01858, N = 3 SE +/- 0.01959, N = 3 7.16575 7.19867 7.20203 MIN: 7.05 MIN: 7.05 MIN: 7.07 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Zstd Compression Compression Level: 19 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 19 Linux 5.4 Linux 5.9 Git Linux 5.8 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.00, N = 3 20.5 20.4 20.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
NCNN Target: CPU - Model: squeezenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: squeezenet Linux 5.8 Linux 5.9 Git Linux 5.4 6 12 18 24 30 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 24.76 24.84 24.88 MIN: 24.65 / MAX: 36.01 MIN: 24.64 / MAX: 36.64 MIN: 24.67 / MAX: 25.89 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Incompact3D Input: Cylinder OpenBenchmarking.org Seconds, Fewer Is Better Incompact3D 2020-09-17 Input: Cylinder Linux 5.4 Linux 5.9 Git Linux 5.8 160 320 480 640 800 SE +/- 1.52, N = 3 SE +/- 0.28, N = 3 SE +/- 0.11, N = 3 750.92 754.27 754.29 1. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
WebP Image Encode Encode Settings: Quality 100, Highest Compression OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Highest Compression Linux 5.4 Linux 5.9 Git Linux 5.8 2 4 6 8 10 SE +/- 0.009, N = 3 SE +/- 0.015, N = 3 SE +/- 0.009, N = 3 8.148 8.161 8.184 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
LAMMPS Molecular Dynamics Simulator Model: Rhodopsin Protein OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 24Aug2020 Model: Rhodopsin Protein Linux 5.9 Git Linux 5.8 Linux 5.4 0.6206 1.2412 1.8618 2.4824 3.103 SE +/- 0.007, N = 3 SE +/- 0.005, N = 3 SE +/- 0.004, N = 3 2.758 2.758 2.746 1. (CXX) g++ options: -O3 -pthread -lm
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.7.3 Encoder Speed: 0 Linux 5.4 Linux 5.8 Linux 5.9 Git 40 80 120 160 200 SE +/- 0.69, N = 3 SE +/- 0.73, N = 3 SE +/- 0.02, N = 3 182.46 182.79 183.23 1. (CXX) g++ options: -O3 -fPIC
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU Linux 5.4 Linux 5.9 Git Linux 5.8 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 18.75 18.79 18.83 MIN: 18.38 MIN: 18.35 MIN: 18.6 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
LibRaw Post-Processing Benchmark OpenBenchmarking.org Mpix/sec, More Is Better LibRaw 0.20 Post-Processing Benchmark Linux 5.4 Linux 5.8 Linux 5.9 Git 7 14 21 28 35 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 27.66 27.65 27.55 1. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm
GROMACS Water Benchmark OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2020.3 Water Benchmark Linux 5.8 Linux 5.4 Linux 5.9 Git 0.1134 0.2268 0.3402 0.4536 0.567 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.004, N = 3 0.504 0.504 0.502 1. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm
FFTE N=256, 3D Complex FFT Routine OpenBenchmarking.org MFLOPS, More Is Better FFTE 7.0 N=256, 3D Complex FFT Routine Linux 5.9 Git Linux 5.4 Linux 5.8 4K 8K 12K 16K 20K SE +/- 64.79, N = 3 SE +/- 95.32, N = 3 SE +/- 59.99, N = 3 20335.31 20328.50 20258.60 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
NCNN Target: CPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mobilenet Linux 5.9 Git Linux 5.8 Linux 5.4 7 14 21 28 35 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 28.84 28.85 28.94 MIN: 28.67 / MAX: 41.4 MIN: 28.7 / MAX: 41.24 MIN: 28.7 / MAX: 32.85 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
AOM AV1 Encoder Mode: Speed 6 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Two-Pass Linux 5.4 Linux 5.9 Git Linux 5.8 0.675 1.35 2.025 2.7 3.375 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 3.00 2.99 2.99 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Monte Carlo Simulations of Ionised Nebulae Input: Dust 2D tau100.0 OpenBenchmarking.org Seconds, Fewer Is Better Monte Carlo Simulations of Ionised Nebulae 2019-03-24 Input: Dust 2D tau100.0 Linux 5.8 Linux 5.9 Git Linux 5.4 70 140 210 280 350 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 304 304 305 1. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O3 -O2 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
NAMD ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms Linux 5.4 Linux 5.8 Linux 5.9 Git 0.8746 1.7492 2.6238 3.4984 4.373 SE +/- 0.00123, N = 3 SE +/- 0.00476, N = 3 SE +/- 0.00417, N = 3 3.87433 3.88624 3.88703
Caffe Model: GoogleNet - Acceleration: CPU - Iterations: 100 OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: GoogleNet - Acceleration: CPU - Iterations: 100 Linux 5.4 Linux 5.8 Linux 5.9 Git 40K 80K 120K 160K 200K SE +/- 195.74, N = 3 SE +/- 11.41, N = 3 SE +/- 205.69, N = 3 164645 165139 165181 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
MPV Video Input: Big Buck Bunny Sunflower 1080p - Decode: Software Only OpenBenchmarking.org FPS, More Is Better MPV Video Input: Big Buck Bunny Sunflower 1080p - Decode: Software Only Linux 5.8 Linux 5.9 Git Linux 5.4 30 60 90 120 150 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 139.88 139.75 139.45 MIN: 123.1 / MAX: 146.35 MIN: 119 / MAX: 146.35 MIN: 115.16 / MAX: 146.34 1. mpv 0.32.0
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: yolov4-tiny Linux 5.8 Linux 5.9 Git Linux 5.4 9 18 27 36 45 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 39.36 39.45 39.48 MIN: 39.23 / MAX: 42.19 MIN: 39.23 / MAX: 42.4 MIN: 39.28 / MAX: 53.47 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
InfluxDB Concurrent Streams: 1024 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 1024 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 Linux 5.4 Linux 5.9 Git Linux 5.8 200K 400K 600K 800K 1000K SE +/- 1038.31, N = 3 SE +/- 2049.30, N = 3 SE +/- 553.23, N = 3 996459.9 994952.9 993437.4
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU Linux 5.9 Git Linux 5.8 Linux 5.4 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 19.88 19.89 19.94 MIN: 19.73 MIN: 19.73 MIN: 19.8 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
AI Benchmark Alpha Device Inference Score OpenBenchmarking.org Score, More Is Better AI Benchmark Alpha 0.1.2 Device Inference Score Linux 5.9 Git Linux 5.4 Linux 5.8 140 280 420 560 700 665 664 663
Caffe Model: AlexNet - Acceleration: CPU - Iterations: 100 OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: AlexNet - Acceleration: CPU - Iterations: 100 Linux 5.9 Git Linux 5.4 Linux 5.8 14K 28K 42K 56K 70K SE +/- 46.28, N = 3 SE +/- 46.52, N = 3 SE +/- 13.57, N = 3 66888 66942 67067 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
OCRMyPDF Processing 60 Page PDF Document OpenBenchmarking.org Seconds, Fewer Is Better OCRMyPDF 9.6.0+dfsg Processing 60 Page PDF Document Linux 5.8 Linux 5.9 Git Linux 5.4 12 24 36 48 60 SE +/- 0.13, N = 3 SE +/- 0.20, N = 3 SE +/- 0.14, N = 3 55.39 55.51 55.54
NCNN Target: CPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet18 Linux 5.9 Git Linux 5.4 Linux 5.8 6 12 18 24 30 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 23.05 23.09 23.11 MIN: 22.9 / MAX: 25.57 MIN: 22.97 / MAX: 26.84 MIN: 22.91 / MAX: 35.18 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Timed LLVM Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 10.0 Time To Compile Linux 5.8 Linux 5.4 Linux 5.9 Git 300 600 900 1200 1500 SE +/- 0.97, N = 3 SE +/- 7.38, N = 3 SE +/- 0.27, N = 3 1490.33 1493.22 1493.85
NCNN Target: CPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: googlenet Linux 5.8 Linux 5.9 Git Linux 5.4 6 12 18 24 30 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 23.03 23.07 23.08 MIN: 22.94 / MAX: 23.49 MIN: 22.94 / MAX: 35.57 MIN: 22.97 / MAX: 26.36 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: shufflenet-v2 Linux 5.8 Linux 5.9 Git Linux 5.4 1.0508 2.1016 3.1524 4.2032 5.254 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 4.66 4.66 4.67 MIN: 4.61 / MAX: 6.54 MIN: 4.59 / MAX: 6.57 MIN: 4.59 / MAX: 8.01 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: BMW27 - Compute: CPU-Only Linux 5.9 Git Linux 5.8 Linux 5.4 80 160 240 320 400 SE +/- 0.53, N = 3 SE +/- 1.11, N = 3 SE +/- 0.56, N = 3 381.93 382.25 382.67
Timed MAFFT Alignment Multiple Sequence Alignment - LSU RNA OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 7.471 Multiple Sequence Alignment - LSU RNA Linux 5.4 Linux 5.8 Linux 5.9 Git 3 6 9 12 15 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.14, N = 3 12.27 12.29 12.29 1. (CC) gcc options: -std=c99 -O3 -lm -lpthread
Timed HMMer Search Pfam Database Search OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.1 Pfam Database Search Linux 5.4 Linux 5.8 Linux 5.9 Git 30 60 90 120 150 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 125.43 125.50 125.66 1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
oneDNN Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU Linux 5.9 Git Linux 5.4 Linux 5.8 2 4 6 8 10 SE +/- 0.00534, N = 3 SE +/- 0.02123, N = 3 SE +/- 0.01758, N = 3 7.40944 7.41923 7.42274 MIN: 7.3 MIN: 7.35 MIN: 7.35 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Kripke OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.4 Linux 5.8 Linux 5.9 Git Linux 5.4 4M 8M 12M 16M 20M SE +/- 59668.65, N = 3 SE +/- 85458.77, N = 3 SE +/- 15901.03, N = 3 17866357 17848217 17835730 1. (CXX) g++ options: -O3 -fopenmp
NCNN Target: CPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU-v3-v3 - Model: mobilenet-v3 Linux 5.9 Git Linux 5.4 Linux 5.8 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 6.52 6.53 6.53 MIN: 6.42 / MAX: 8.56 MIN: 6.41 / MAX: 9.17 MIN: 6.43 / MAX: 9.7 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
ASTC Encoder Preset: Thorough OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Thorough Linux 5.4 Linux 5.8 Linux 5.9 Git 16 32 48 64 80 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 72.26 72.32 72.37 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
AI Benchmark Alpha Device AI Score OpenBenchmarking.org Score, More Is Better AI Benchmark Alpha 0.1.2 Device AI Score Linux 5.9 Git Linux 5.4 Linux 5.8 300 600 900 1200 1500 1398 1397 1396
Zstd Compression Compression Level: 3 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 3 Linux 5.8 Linux 5.9 Git Linux 5.4 500 1000 1500 2000 2500 SE +/- 1.97, N = 3 SE +/- 1.34, N = 3 SE +/- 6.19, N = 3 2389.3 2389.0 2386.0 1. (CC) gcc options: -O3 -pthread -lz -llzma
TensorFlow Lite Model: Mobilenet Quant OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Mobilenet Quant Linux 5.8 Linux 5.4 Linux 5.9 Git 70K 140K 210K 280K 350K SE +/- 64.01, N = 3 SE +/- 72.59, N = 3 SE +/- 102.05, N = 3 348976 348999 349438
Hierarchical INTegration Test: FLOAT OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT Linux 5.4 Linux 5.8 Linux 5.9 Git 80M 160M 240M 320M 400M SE +/- 1362924.61, N = 3 SE +/- 293442.33, N = 3 SE +/- 715889.13, N = 3 381569301.13 381270686.12 381134357.52 1. (CC) gcc options: -O3 -march=native -lm
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Fishy Cat - Compute: CPU-Only Linux 5.8 Linux 5.9 Git Linux 5.4 120 240 360 480 600 SE +/- 0.55, N = 3 SE +/- 0.81, N = 3 SE +/- 0.85, N = 3 531.91 532.18 532.49
Rodinia Test: OpenMP Streamcluster OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Streamcluster Linux 5.9 Git Linux 5.8 Linux 5.4 7 14 21 28 35 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 28.70 28.73 28.73 1. (CXX) g++ options: -O2 -lOpenCL
NCNN Target: CPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet50 Linux 5.8 Linux 5.9 Git Linux 5.4 11 22 33 44 55 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 46.45 46.46 46.50 MIN: 46.3 / MAX: 49.28 MIN: 46.33 / MAX: 51.2 MIN: 46.31 / MAX: 58.67 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
TNN Target: CPU - Model: MobileNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: MobileNet v2 Linux 5.4 Linux 5.8 Linux 5.9 Git 80 160 240 320 400 SE +/- 0.47, N = 3 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 366.30 366.39 366.63 MIN: 364.37 / MAX: 374.6 MIN: 365.21 / MAX: 367.87 MIN: 365.56 / MAX: 368.1 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
ASTC Encoder Preset: Medium OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Medium Linux 5.4 Linux 5.8 Linux 5.9 Git 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 11.14 11.14 11.15 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Caffe Model: AlexNet - Acceleration: CPU - Iterations: 200 OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: AlexNet - Acceleration: CPU - Iterations: 200 Linux 5.9 Git Linux 5.4 Linux 5.8 30K 60K 90K 120K 150K SE +/- 97.27, N = 3 SE +/- 12.01, N = 3 SE +/- 43.43, N = 3 133774 133823 133870 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
TensorFlow Lite Model: NASNet Mobile OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: NASNet Mobile Linux 5.8 Linux 5.4 Linux 5.9 Git 80K 160K 240K 320K 400K SE +/- 108.52, N = 3 SE +/- 60.60, N = 3 SE +/- 77.03, N = 3 366914 366947 367078
TensorFlow Lite Model: SqueezeNet OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: SqueezeNet Linux 5.4 Linux 5.9 Git Linux 5.8 110K 220K 330K 440K 550K SE +/- 34.51, N = 3 SE +/- 63.00, N = 3 SE +/- 53.76, N = 3 499694 499825 499845
ASTC Encoder Preset: Exhaustive OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Exhaustive Linux 5.4 Linux 5.9 Git Linux 5.8 120 240 360 480 600 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.22, N = 3 578.17 578.22 578.30 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
TensorFlow Lite Model: Inception ResNet V2 OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Inception ResNet V2 Linux 5.4 Linux 5.8 Linux 5.9 Git 1.4M 2.8M 4.2M 5.6M 7M SE +/- 634.06, N = 3 SE +/- 88.19, N = 3 SE +/- 862.41, N = 3 6517420 6518137 6518793
TensorFlow Lite Model: Inception V4 OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Inception V4 Linux 5.9 Git Linux 5.4 Linux 5.8 1.5M 3M 4.5M 6M 7.5M SE +/- 580.35, N = 3 SE +/- 250.95, N = 3 SE +/- 132.46, N = 3 7202907 7204167 7204273
TensorFlow Lite Model: Mobilenet Float OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Mobilenet Float Linux 5.8 Linux 5.4 Linux 5.9 Git 70K 140K 210K 280K 350K SE +/- 39.28, N = 3 SE +/- 59.09, N = 3 SE +/- 48.17, N = 3 339372 339412 339434
AI Benchmark Alpha Device Training Score OpenBenchmarking.org Score, More Is Better AI Benchmark Alpha 0.1.2 Device Training Score Linux 5.9 Git Linux 5.8 Linux 5.4 160 320 480 640 800 733 733 733
AOM AV1 Encoder Mode: Speed 6 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Realtime Linux 5.9 Git Linux 5.8 Linux 5.4 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 14.99 14.99 14.99 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 0 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 0 Two-Pass Linux 5.9 Git Linux 5.8 Linux 5.4 0.0495 0.099 0.1485 0.198 0.2475 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.22 0.22 0.22 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
InfluxDB Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 Linux 5.4 Linux 5.8 Linux 5.9 Git 200K 400K 600K 800K 1000K SE +/- 11259.18, N = 12 SE +/- 19326.59, N = 12 SE +/- 21169.01, N = 12 870964.9 814462.2 805704.5
Mlpack Benchmark Benchmark: scikit_qda OpenBenchmarking.org Seconds, Fewer Is Better Mlpack Benchmark Benchmark: scikit_qda Linux 5.8 Linux 5.9 Git Linux 5.4 20 40 60 80 100 SE +/- 1.29, N = 3 SE +/- 1.27, N = 12 SE +/- 1.98, N = 12 84.76 86.12 88.35
NCNN Target: Vulkan GPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: blazeface Linux 5.9 Git Linux 5.8 Linux 5.4 0.5918 1.1836 1.7754 2.3672 2.959 SE +/- 0.03, N = 3 SE +/- 0.15, N = 3 SE +/- 0.01, N = 3 2.29 2.35 2.63 MIN: 2.14 / MAX: 2.48 MIN: 2.17 / MAX: 2.76 MIN: 2.59 / MAX: 2.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: shufflenet-v2 Linux 5.8 Linux 5.4 Linux 5.9 Git 3 6 9 12 15 SE +/- 0.30, N = 3 SE +/- 0.01, N = 3 SE +/- 0.52, N = 3 10.06 10.11 11.09 MIN: 9.19 / MAX: 10.9 MIN: 9.97 / MAX: 10.55 MIN: 10.03 / MAX: 13.96 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Phoronix Test Suite v10.8.4