EPYC 7502 Xmas Week

AMD EPYC 7502 32-Core testing with a ASRockRack EPYCD8 (P2.10 BIOS) and llvmpipe on Ubuntu 20.10 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2012205-HA-EPYC7502X40.

EPYC 7502 Xmas WeekProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionAMD EPYC 750223AMD EPYC 7502 32-Core @ 2.50GHz (32 Cores / 64 Threads)ASRockRack EPYCD8 (P2.10 BIOS)AMD Starship/Matisse126GB280GB INTEL SSDPED1D280GAllvmpipeAMD Starship/MatisseVE2282 x Intel I350Ubuntu 20.105.8.0-31-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 10.2.0ext41920x1080OpenBenchmarking.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++,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 Processor Details- Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x830101c Security 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 7502 Xmas Weekclomp: Static OMP Speedupsimdjson: Kostyasimdjson: LargeRandsimdjson: PartialTweetssimdjson: DistinctUserIDonednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUbuild2: Time To Compilebuild-eigen: Time To Compileencode-ape: WAV To APEnode-web-tooling: ncnn: 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-tinyncnn: CPU - squeezenet_ssdncnn: CPU - regnety_400mencode-wavpack: WAV To WavPackAMD EPYC 75022364.40.510.340.570.581.785003.292781.278911.108064.116242.359553.832944.358802.373482.076523205.711136.513227.051155.990.5895313223.811160.031.2580286.08099.85414.4567.6926.0011.5110.6411.2410.5714.485.3524.4839.8316.6510.4731.1433.5328.3271.1116.06767.80.510.330.570.581.801693.369611.281021.115204.073112.337183.842094.394282.370252.074633182.891152.553242.451153.350.5910163201.101178.841.2615986.15199.07914.4807.7024.9611.5810.7211.3610.6415.185.2925.8439.7817.3711.1131.4234.1428.0871.0716.08365.30.510.330.570.581.780683.284831.274931.077674.111942.336273.790064.391192.378182.069083210.611172.013208.361176.180.5853623216.211167.551.2665186.37499.12314.4237.7725.1411.7510.7711.3210.8615.155.2725.4340.1017.2011.0931.1934.0627.6270.4516.072OpenBenchmarking.org

CLOMP

Static OMP Speedup

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP SpeedupAMD EPYC 7502231530456075SE +/- 0.59, N = 3SE +/- 0.83, N = 5SE +/- 1.04, N = 364.467.865.31. (CC) gcc options: -fopenmp -O3 -lm

simdjson

Throughput Test: Kostya

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: KostyaAMD EPYC 7502230.11480.22960.34440.45920.574SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.510.510.511. (CXX) g++ options: -O3 -pthread

simdjson

Throughput Test: LargeRandom

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandomAMD EPYC 7502230.07650.1530.22950.3060.3825SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.340.330.331. (CXX) g++ options: -O3 -pthread

simdjson

Throughput Test: PartialTweets

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: PartialTweetsAMD EPYC 7502230.12830.25660.38490.51320.6415SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.570.570.571. (CXX) g++ options: -O3 -pthread

simdjson

Throughput Test: DistinctUserID

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: DistinctUserIDAMD EPYC 7502230.13050.2610.39150.5220.6525SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 50.580.580.581. (CXX) g++ options: -O3 -pthread

oneDNN

Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUAMD EPYC 7502230.40540.81081.21621.62162.027SE +/- 0.00953, N = 3SE +/- 0.00485, N = 3SE +/- 0.02001, N = 31.785001.801691.78068MIN: 1.52MIN: 1.52MIN: 1.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUAMD EPYC 7502230.75821.51642.27463.03283.791SE +/- 0.01054, N = 3SE +/- 0.03752, N = 3SE +/- 0.02370, N = 33.292783.369613.28483MIN: 2.89MIN: 2.94MIN: 2.881. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7502230.28820.57640.86461.15281.441SE +/- 0.00427, N = 3SE +/- 0.00229, N = 3SE +/- 0.00289, N = 31.278911.281021.27493MIN: 1.13MIN: 1.13MIN: 1.131. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7502230.25090.50180.75271.00361.2545SE +/- 0.01806, N = 3SE +/- 0.00829, N = 3SE +/- 0.00488, N = 31.108061.115201.07767MIN: 0.95MIN: 0.95MIN: 0.931. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUAMD EPYC 7502230.92621.85242.77863.70484.631SE +/- 0.00333, N = 3SE +/- 0.01990, N = 3SE +/- 0.05263, N = 34.116244.073114.11194MIN: 3.35MIN: 3.38MIN: 3.341. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUAMD EPYC 7502230.53091.06181.59272.12362.6545SE +/- 0.01807, N = 3SE +/- 0.00438, N = 3SE +/- 0.00576, N = 32.359552.337182.33627MIN: 2.07MIN: 2.06MIN: 2.051. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUAMD EPYC 7502230.86451.7292.59353.4584.3225SE +/- 0.01461, N = 3SE +/- 0.04927, N = 3SE +/- 0.01436, N = 33.832943.842093.79006MIN: 3.49MIN: 3.52MIN: 3.511. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7502230.98871.97742.96613.95484.9435SE +/- 0.02731, N = 3SE +/- 0.01475, N = 3SE +/- 0.01128, N = 34.358804.394284.39119MIN: 3.89MIN: 3.89MIN: 3.871. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7502230.53511.07021.60532.14042.6755SE +/- 0.00748, N = 3SE +/- 0.00493, N = 3SE +/- 0.01179, N = 32.373482.370252.37818MIN: 2.08MIN: 2.07MIN: 2.081. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7502230.46720.93441.40161.86882.336SE +/- 0.03185, N = 3SE +/- 0.00900, N = 3SE +/- 0.00643, N = 32.076522.074632.06908MIN: 1.87MIN: 1.85MIN: 1.861. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUAMD EPYC 7502237001400210028003500SE +/- 4.27, N = 3SE +/- 8.61, N = 3SE +/- 12.01, N = 33205.713182.893210.61MIN: 3088.13MIN: 3043.08MIN: 3095.231. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUAMD EPYC 75022330060090012001500SE +/- 4.77, N = 3SE +/- 8.23, N = 3SE +/- 6.06, N = 31136.511152.551172.01MIN: 1054.18MIN: 1088.54MIN: 1086.641. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7502237001400210028003500SE +/- 13.02, N = 3SE +/- 26.07, N = 3SE +/- 11.53, N = 33227.053242.453208.36MIN: 3076.05MIN: 3103.97MIN: 3078.811. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUAMD EPYC 75022330060090012001500SE +/- 4.70, N = 3SE +/- 6.97, N = 3SE +/- 5.46, N = 31155.991153.351176.18MIN: 1062.75MIN: 1073.9MIN: 1101.031. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUAMD EPYC 7502230.1330.2660.3990.5320.665SE +/- 0.000625, N = 3SE +/- 0.001851, N = 3SE +/- 0.000678, N = 30.5895310.5910160.585362MIN: 0.52MIN: 0.52MIN: 0.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUAMD EPYC 7502237001400210028003500SE +/- 6.02, N = 3SE +/- 8.44, N = 3SE +/- 10.71, N = 33223.813201.103216.21MIN: 3103.68MIN: 3085.77MIN: 3096.381. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUAMD EPYC 75022330060090012001500SE +/- 16.25, N = 3SE +/- 6.76, N = 3SE +/- 7.10, N = 31160.031178.841167.55MIN: 1081.43MIN: 1109.11MIN: 1086.31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUAMD EPYC 7502230.2850.570.8551.141.425SE +/- 0.00653, N = 3SE +/- 0.00063, N = 3SE +/- 0.00287, N = 31.258021.261591.26651MIN: 1.07MIN: 1.07MIN: 1.071. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Build2

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To CompileAMD EPYC 75022320406080100SE +/- 0.11, N = 3SE +/- 0.19, N = 3SE +/- 0.06, N = 386.0886.1586.37

Timed Eigen Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To CompileAMD EPYC 75022320406080100SE +/- 0.30, N = 3SE +/- 0.00, N = 3SE +/- 0.13, N = 399.8599.0899.12

Monkey Audio Encoding

WAV To APE

OpenBenchmarking.orgSeconds, Fewer Is BetterMonkey Audio Encoding 3.99.6WAV To APEAMD EPYC 75022348121620SE +/- 0.03, N = 5SE +/- 0.02, N = 5SE +/- 0.02, N = 514.4614.4814.421. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt

Node.js V8 Web Tooling Benchmark

OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling BenchmarkAMD EPYC 750223246810SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 37.697.707.771. Nodejs v12.18.2

NCNN

Target: CPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenetAMD EPYC 750223612182430SE +/- 0.35, N = 3SE +/- 0.21, N = 15SE +/- 0.25, N = 1526.0024.9625.14MIN: 23.4 / MAX: 80.59MIN: 21.51 / MAX: 105.49MIN: 21.73 / MAX: 94.131. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v2AMD EPYC 7502233691215SE +/- 0.30, N = 3SE +/- 0.11, N = 15SE +/- 0.17, N = 1511.5111.5811.75MIN: 9.54 / MAX: 78.57MIN: 9.53 / MAX: 160.97MIN: 9.44 / MAX: 142.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3AMD EPYC 7502233691215SE +/- 0.32, N = 3SE +/- 0.12, N = 15SE +/- 0.13, N = 1510.6410.7210.77MIN: 8.92 / MAX: 145.19MIN: 8.74 / MAX: 196.77MIN: 8.87 / MAX: 120.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v2AMD EPYC 7502233691215SE +/- 0.05, N = 3SE +/- 0.09, N = 15SE +/- 0.09, N = 1511.2411.3611.32MIN: 9.67 / MAX: 30.09MIN: 9.54 / MAX: 133.63MIN: 9.56 / MAX: 126.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnetAMD EPYC 7502233691215SE +/- 0.19, N = 3SE +/- 0.09, N = 15SE +/- 0.17, N = 1510.5710.6410.86MIN: 8.64 / MAX: 110.66MIN: 8.68 / MAX: 98.56MIN: 8.82 / MAX: 109.381. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b0AMD EPYC 75022348121620SE +/- 0.01, N = 3SE +/- 0.17, N = 15SE +/- 0.11, N = 1514.4815.1815.15MIN: 12.55 / MAX: 38.15MIN: 12.52 / MAX: 114.66MIN: 12.72 / MAX: 99.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazefaceAMD EPYC 7502231.20382.40763.61144.81526.019SE +/- 0.27, N = 3SE +/- 0.04, N = 15SE +/- 0.07, N = 155.355.295.27MIN: 4.5 / MAX: 76.73MIN: 4.5 / MAX: 68.13MIN: 4.45 / MAX: 84.291. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenetAMD EPYC 750223612182430SE +/- 0.79, N = 3SE +/- 0.45, N = 15SE +/- 0.38, N = 1424.4825.8425.43MIN: 20.77 / MAX: 294.45MIN: 20.85 / MAX: 143.14MIN: 21.61 / MAX: 125.031. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg16AMD EPYC 750223918273645SE +/- 0.37, N = 3SE +/- 0.27, N = 15SE +/- 0.29, N = 1539.8339.7840.10MIN: 34.39 / MAX: 142.77MIN: 33.62 / MAX: 143.49MIN: 33.09 / MAX: 166.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18AMD EPYC 75022348121620SE +/- 0.32, N = 3SE +/- 0.15, N = 15SE +/- 0.17, N = 1516.6517.3717.20MIN: 13.39 / MAX: 102.22MIN: 13.34 / MAX: 144.37MIN: 13.21 / MAX: 165.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnetAMD EPYC 7502233691215SE +/- 0.16, N = 3SE +/- 0.16, N = 15SE +/- 0.15, N = 1510.4711.1111.09MIN: 8.95 / MAX: 25.75MIN: 8.31 / MAX: 133.74MIN: 8.7 / MAX: 147.451. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50AMD EPYC 750223714212835SE +/- 0.56, N = 3SE +/- 0.25, N = 15SE +/- 0.29, N = 1531.1431.4231.19MIN: 26.59 / MAX: 151.25MIN: 26.5 / MAX: 221.62MIN: 26.19 / MAX: 194.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tinyAMD EPYC 750223816243240SE +/- 0.52, N = 3SE +/- 0.21, N = 15SE +/- 0.26, N = 1533.5334.1434.06MIN: 29.02 / MAX: 105.06MIN: 29.19 / MAX: 174.89MIN: 28.7 / MAX: 151.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssdAMD EPYC 750223714212835SE +/- 0.09, N = 3SE +/- 0.18, N = 15SE +/- 0.19, N = 1528.3228.0827.62MIN: 25.19 / MAX: 156.92MIN: 24.38 / MAX: 175.25MIN: 24.14 / MAX: 2081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400mAMD EPYC 7502231632486480SE +/- 2.12, N = 3SE +/- 0.31, N = 15SE +/- 0.34, N = 1571.1171.0770.45MIN: 63.73 / MAX: 270.85MIN: 63.46 / MAX: 200.04MIN: 63 / MAX: 192.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

WavPack Audio Encoding

WAV To WavPack

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPackAMD EPYC 75022348121620SE +/- 0.00, N = 5SE +/- 0.01, N = 5SE +/- 0.00, N = 516.0716.0816.071. (CXX) g++ options: -rdynamic

Phoronix Test Suite v10.8.4