EPYC Linux 5.9

2 x AMD EPYC 7272 12-Core testing with a Supermicro H11DSi-NT v2.00 (2.1 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2009224-FI-EPYCLINUX47.

EPYC Linux 5.9ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen ResolutionLinux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+2 x AMD EPYC 7272 12-Core @ 2.90GHz (24 Cores / 48 Threads)Supermicro H11DSi-NT v2.00 (2.1 BIOS)AMD Starship/Matisse504GB280GB INTEL SSDPE21D280GAASPEEDVE2282 x Intel 10G X550TUbuntu 20.045.8.0-050800rc6daily20200721-generic (x86_64) 20200720GNOME Shell 3.36.1X Server 1.20.8modesetting 1.20.8GCC 9.3.0ext41920x10805.8.10-050810-generic (x86_64)5.9.0-050900rc6daily20200922-generic (x86_64) 20200921OpenBenchmarking.orgCompiler 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: acpi-cpufreq ondemand - CPU Microcode: 0x8301034Python Details- Python 2.7.18rc1 + Python 3.8.2Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

EPYC Linux 5.9incompact3d: Cylindergpaw: Carbon Nanotubemnn: SqueezeNetV1.0mnn: resnet-v2-50mnn: MobileNetV2_224mnn: mobilenet-v1-1.0mnn: inception-v3ncnn: CPU - squeezenet_int8ncnn: CPU - mobilenet_v3ncnn: CPU - squeezenetncnn: CPU - mnasnetncnn: CPU - blazefacencnn: CPU - googlenet_int8ncnn: CPU - vgg16_int8ncnn: CPU - resnet18_int8ncnn: CPU - alexnetncnn: CPU - resnet50_int8ncnn: CPU - mobilenetv2_yolov3kripke: Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+224.495275113.23013.14640.21813.9689.32543.01325.9218.1322.6521.727.8969.27120.9535.2312.1993.8029.8661873427222.295451114.47014.97640.83115.85610.12642.84926.9616.2421.8225.858.7584.66127.4735.8812.2092.4932.1462763517220.988174113.51113.98640.09014.24510.81243.44928.0617.3924.1522.188.4470.84118.2036.1712.3192.3330.6863306793OpenBenchmarking.org

Incompact3D

Input: Cylinder

OpenBenchmarking.orgSeconds, Fewer Is BetterIncompact3D 2020-09-17Input: CylinderLinux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+50100150200250SE +/- 0.65, N = 3SE +/- 0.38, N = 3SE +/- 0.34, N = 3224.50222.30220.991. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

GPAW

Input: Carbon Nanotube

OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 20.1Input: Carbon NanotubeLinux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+306090120150SE +/- 0.36, N = 3SE +/- 0.12, N = 3SE +/- 0.50, N = 3113.23114.47113.511. (CC) gcc options: -pthread -shared -fwrapv -O2 -lxc -lblas -lmpi

Mobile Neural Network

Model: SqueezeNetV1.0

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: SqueezeNetV1.0Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+48121620SE +/- 0.41, N = 12SE +/- 0.67, N = 9SE +/- 0.24, N = 1213.1514.9813.99MIN: 10.52 / MAX: 20.05MIN: 11.65 / MAX: 23MIN: 10.28 / MAX: 19.151. (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

Mobile Neural Network

Model: resnet-v2-50

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: resnet-v2-50Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+918273645SE +/- 0.64, N = 12SE +/- 0.64, N = 9SE +/- 0.47, N = 1240.2240.8340.09MIN: 33.55 / MAX: 150.88MIN: 34.58 / MAX: 136.91MIN: 34.67 / MAX: 136.291. (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

Mobile Neural Network

Model: MobileNetV2_224

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: MobileNetV2_224Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+48121620SE +/- 1.28, N = 12SE +/- 1.31, N = 9SE +/- 1.16, N = 1213.9715.8614.25MIN: 7.57 / MAX: 27.35MIN: 6.77 / MAX: 27.9MIN: 7.55 / MAX: 31.661. (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

Mobile Neural Network

Model: mobilenet-v1-1.0

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: mobilenet-v1-1.0Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+3691215SE +/- 0.625, N = 12SE +/- 0.719, N = 9SE +/- 0.637, N = 129.32510.12610.812MIN: 6.45 / MAX: 22.74MIN: 6.32 / MAX: 21.92MIN: 6.53 / MAX: 17.551. (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

Mobile Neural Network

Model: inception-v3

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2020-09-17Model: inception-v3Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+1020304050SE +/- 0.32, N = 12SE +/- 0.43, N = 9SE +/- 0.29, N = 1243.0142.8543.45MIN: 37.29 / MAX: 208.57MIN: 37.24 / MAX: 265.28MIN: 38.41 / MAX: 235.341. (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: squeezenet_int8

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: squeezenet_int8Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+714212835SE +/- 0.40, N = 12SE +/- 0.45, N = 3SE +/- 1.12, N = 1225.9226.9628.06MIN: 23.3 / MAX: 34.26MIN: 23.61 / MAX: 42.52MIN: 23.38 / MAX: 95.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mobilenet_v3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mobilenet_v3Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+48121620SE +/- 1.14, N = 12SE +/- 4.39, N = 3SE +/- 1.82, N = 1218.1316.2417.39MIN: 10.45 / MAX: 52.44MIN: 11.22 / MAX: 29.7MIN: 10.52 / MAX: 215.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: squeezenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: squeezenetLinux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+612182430SE +/- 0.56, N = 12SE +/- 0.98, N = 3SE +/- 1.28, N = 1222.6521.8224.15MIN: 13.09 / MAX: 44.24MIN: 16.97 / MAX: 34.26MIN: 13.88 / MAX: 67.031. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mnasnetLinux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+612182430SE +/- 1.67, N = 12SE +/- 1.49, N = 3SE +/- 1.69, N = 1221.7225.8522.18MIN: 10.48 / MAX: 61.29MIN: 10.53 / MAX: 34.21MIN: 10.29 / MAX: 51.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: blazefaceLinux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+246810SE +/- 0.30, N = 12SE +/- 0.50, N = 3SE +/- 0.70, N = 127.898.758.44MIN: 4.75 / MAX: 11.08MIN: 4.95 / MAX: 10.07MIN: 5.02 / MAX: 579.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: googlenet_int8

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: googlenet_int8Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+20406080100SE +/- 0.27, N = 12SE +/- 13.75, N = 3SE +/- 1.30, N = 1269.2784.6670.84MIN: 63.84 / MAX: 187.59MIN: 64.86 / MAX: 5374.31MIN: 64.63 / MAX: 3425.611. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vgg16_int8

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: vgg16_int8Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+306090120150SE +/- 1.21, N = 12SE +/- 1.50, N = 3SE +/- 1.19, N = 12120.95127.47118.20MIN: 93.52 / MAX: 1068.39MIN: 98.15 / MAX: 398.17MIN: 93.13 / MAX: 679.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet18_int8

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet18_int8Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+816243240SE +/- 0.69, N = 12SE +/- 1.36, N = 3SE +/- 1.24, N = 1235.2335.8836.17MIN: 31.44 / MAX: 139.23MIN: 30.74 / MAX: 80.83MIN: 30.84 / MAX: 340.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: alexnetLinux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+3691215SE +/- 0.04, N = 12SE +/- 0.11, N = 3SE +/- 0.08, N = 1212.1912.2012.31MIN: 11.84 / MAX: 43.48MIN: 11.91 / MAX: 14.72MIN: 11.86 / MAX: 15.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet50_int8

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet50_int8Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+20406080100SE +/- 0.44, N = 12SE +/- 0.01, N = 3SE +/- 0.37, N = 1293.8092.4992.33MIN: 90.04 / MAX: 676.56MIN: 89.9 / MAX: 181.85MIN: 87.78 / MAX: 213.131. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mobilenetv2_yolov3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mobilenetv2_yolov3Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+714212835SE +/- 1.29, N = 12SE +/- 4.35, N = 3SE +/- 1.42, N = 1229.8632.1430.68MIN: 25.04 / MAX: 62.75MIN: 25.31 / MAX: 48.23MIN: 25.26 / MAX: 2484.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Kripke

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.4Linux 5.8-rc6Linux 5.8.10Linux 5.9-rc6+14M28M42M56M70MSE +/- 855690.21, N = 3SE +/- 1011103.97, N = 3SE +/- 457622.40, N = 36187342762763517633067931. (CXX) g++ options: -O3 -fopenmp

Phoronix Test Suite v10.8.4