EPYC 7F72 NCNN

AMD EPYC 7F72 24-Core testing with a ASRockRack EPYCD8 (P2.10 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2009244-FI-EPYC7F72N05.

EPYC 7F72 NCNNProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen Resolution123AMD EPYC 7F72 24-Core @ 3.20GHz (24 Cores / 48 Threads)ASRockRack EPYCD8 (P2.10 BIOS)AMD Starship/Matisse126GB3841GB Micron_9300_MTFDHAL3T8TDPASPEEDAMD Starship/Matisse2 x Intel I350Ubuntu 20.045.9.0-050900rc6daily20200921-generic (x86_64) 20200920GNOME Shell 3.36.4X Server 1.20.8modesetting 1.20.8GCC 9.3.0ext41024x768OpenBenchmarking.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: 0x830101cSecurity 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: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

EPYC 7F72 NCNNncnn: CPU - squeezenetncnn: CPU - mobilenetncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - shufflenet-v2ncnn: CPU - mnasnetncnn: CPU - efficientnet-b0ncnn: CPU - blazefacencnn: CPU - googlenetncnn: CPU - vgg16ncnn: CPU - resnet18ncnn: CPU - alexnetncnn: CPU - resnet50ncnn: CPU - yolov4-tiny12343.3720.849.258.749.198.3610.983.7548.1473.2824.869.0066.7129.6042.4819.929.108.689.308.3510.963.7248.2373.8924.869.9266.6430.2642.9519.859.158.599.128.2510.853.7048.2175.4224.9110.2366.9530.19OpenBenchmarking.org

NCNN

Target: CPU - Model: squeezenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: squeezenet1231020304050SE +/- 0.35, N = 3SE +/- 0.38, N = 3SE +/- 0.70, N = 343.3742.4842.95MIN: 41.55 / MAX: 143.63MIN: 41.32 / MAX: 81.03MIN: 41.16 / MAX: 48.981. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mobilenet123510152025SE +/- 0.12, N = 3SE +/- 0.63, N = 3SE +/- 0.30, N = 320.8419.9219.85MIN: 20.13 / MAX: 22.95MIN: 18.7 / MAX: 22.04MIN: 18.88 / MAX: 21.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU-v2-v2 - Model: mobilenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU-v2-v2 - Model: mobilenet-v21233691215SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.19, N = 39.259.109.15MIN: 8.84 / MAX: 13.16MIN: 8.74 / MAX: 10.49MIN: 8.66 / MAX: 78.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU-v3-v3 - Model: mobilenet-v3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU-v3-v3 - Model: mobilenet-v3123246810SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 38.748.688.59MIN: 8.3 / MAX: 10.01MIN: 8.39 / MAX: 10.64MIN: 8.28 / MAX: 10.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: shufflenet-v21233691215SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 39.199.309.12MIN: 9.03 / MAX: 10.91MIN: 8.98 / MAX: 81.47MIN: 8.9 / MAX: 10.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: mnasnet123246810SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 38.368.358.25MIN: 7.78 / MAX: 9.83MIN: 7.75 / MAX: 10.14MIN: 7.83 / MAX: 101. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: efficientnet-b01233691215SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.08, N = 310.9810.9610.85MIN: 10.59 / MAX: 28.76MIN: 10.61 / MAX: 12.94MIN: 10.61 / MAX: 14.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: blazeface1230.84381.68762.53143.37524.219SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 33.753.723.70MIN: 3.5 / MAX: 10.48MIN: 3.54 / MAX: 3.88MIN: 3.5 / MAX: 9.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: googlenet1231122334455SE +/- 0.30, N = 3SE +/- 0.24, N = 3SE +/- 0.03, N = 348.1448.2348.21MIN: 46.95 / MAX: 145.26MIN: 46.99 / MAX: 156.56MIN: 47.1 / MAX: 53.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: vgg1612320406080100SE +/- 1.15, N = 3SE +/- 0.26, N = 3SE +/- 0.38, N = 373.2873.8975.42MIN: 69.31 / MAX: 264.97MIN: 71.19 / MAX: 118.48MIN: 71.81 / MAX: 93.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet18123612182430SE +/- 0.39, N = 3SE +/- 0.50, N = 3SE +/- 0.55, N = 324.8624.8624.91MIN: 24.07 / MAX: 116MIN: 24.09 / MAX: 26.16MIN: 24.1 / MAX: 27.431. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: alexnet1233691215SE +/- 0.15, N = 3SE +/- 0.24, N = 3SE +/- 0.11, N = 39.009.9210.23MIN: 8.58 / MAX: 10.93MIN: 9.32 / MAX: 12.38MIN: 9.88 / MAX: 12.211. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: resnet501231530456075SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.54, N = 366.7166.6466.95MIN: 66.24 / MAX: 68.87MIN: 66.22 / MAX: 68.65MIN: 65.79 / MAX: 139.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: CPU - Model: yolov4-tiny123714212835SE +/- 0.11, N = 3SE +/- 0.16, N = 3SE +/- 0.12, N = 329.6030.2630.19MIN: 29.02 / MAX: 31.67MIN: 29.46 / MAX: 83.42MIN: 29.59 / MAX: 33.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Phoronix Test Suite v10.8.4