EPYC 7601 Extra

AMD EPYC 7601 32-Core testing with a ASRockRack EPYCD8 (P2.10 BIOS) and llvmpipe 126GB on Ubuntu 19.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1911029-HU-EPYC7601E78
Jump To Table - Results

Statistics

Remove Outliers Before Calculating Averages

Graph Settings

Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
EPYC 7601
November 01 2019
  4 Hours, 37 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

EPYC 7601

Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: ASRockRack EPYCD8 (P2.10 BIOS), Chipset: AMD 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPED1D280GA, Graphics: llvmpipe 126GB, Network: 2 x Intel I350

OS: Ubuntu 19.10, Kernel: 5.4.0-999-generic (x86_64) 20191030, Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 3.3 Mesa 19.2.1 (LLVM 9.0 128 bits), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 1024x768

Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-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 Notes: Scaling Governor: acpi-cpufreq ondemand
Security Notes: 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 STIBP: disabled RSB filling

EPYC 7601 ExtraOpenBenchmarking.orgPhoronix Test SuiteAMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads)ASRockRack EPYCD8 (P2.10 BIOS)AMD 17h129024MB280GB INTEL SSDPED1D280GAllvmpipe 126GB2 x Intel I350Ubuntu 19.105.4.0-999-generic (x86_64) 20191030GNOME Shell 3.34.1X Server 1.20.5modesetting 1.20.53.3 Mesa 19.2.1 (LLVM 9.0 128 bits)GCC 9.2.1 20191008ext41024x768ProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionEPYC 7601 Extra BenchmarksSystem Logs- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-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 - Scaling Governor: acpi-cpufreq ondemand- 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 STIBP: disabled RSB filling

EPYC 7601 Extrasvt-av1: Enc Mode 0 - 1080pmkl-dnn: Convolution Batch conv_all - u8s8f32mkl-dnn: Convolution Batch conv_all - f32fftw: Float + SSE - 2D FFT Size 4096mkl-dnn: Deconvolution Batch deconv_all - f32dav1d: Chimera 1080p 10-bitfftw: Stock - 2D FFT Size 4096ospray: San Miguel - SciVisospray: San Miguel - Path Tracermkl-dnn: Convolution Batch conv_googlenet_v3 - u8s8f32mkl-dnn: Convolution Batch conv_googlenet_v3 - f32fftw: Float + SSE - 2D FFT Size 2048ospray: XFrog Forest - Path Tracermkl-dnn: IP Batch 1D - f32fftw: Stock - 2D FFT Size 2048ospray: XFrog Forest - SciVismkl-dnn: Convolution Batch conv_3d - u8s8f32embree: Pathtracer ISPC - Asian Dragon Objmkl-dnn: IP Batch All - f32mkl-dnn: IP Batch All - u8s8f32embree: Pathtracer - Asian Dragon Objmkl-dnn: Deconvolution Batch deconv_1d - f32ospray: NASA Streamlines - SciVismkl-dnn: Convolution Batch conv_3d - f32mkl-dnn: Recurrent Neural Network Training - f32ospray: NASA Streamlines - Path Tracerdav1d: Chimera 1080pembree: Pathtracer ISPC - Crownembree: Pathtracer - Crownfftw: Float + SSE - 2D FFT Size 1024embree: Pathtracer ISPC - Asian Dragonembree: Pathtracer - Asian Dragonmkl-dnn: Deconvolution Batch deconv_1d - u8s8f32mkl-dnn: Convolution Batch conv_alexnet - u8s8f32dav1d: Summer Nature 4Ksvt-av1: Enc Mode 4 - 1080pospray: Magnetic Reconnection - SciVismkl-dnn: Deconvolution Batch deconv_3d - u8s8f32mkl-dnn: Convolution Batch conv_alexnet - f32fftw: Stock - 2D FFT Size 1024mkl-dnn: Deconvolution Batch deconv_3d - f32mkl-dnn: IP Batch 1D - u8s8f32oidn: Memorialsvt-av1: Enc Mode 8 - 1080pdav1d: Summer Nature 1080pfftw: Float + SSE - 1D FFT Size 4096fftw: Float + SSE - 2D FFT Size 512fftw: Stock - 1D FFT Size 4096fftw: Float + SSE - 1D FFT Size 2048fftw: Stock - 1D FFT Size 2048fftw: Stock - 2D FFT Size 512fftw: Float + SSE - 1D FFT Size 1024fftw: Float + SSE - 2D FFT Size 256fftw: Stock - 1D FFT Size 1024fftw: Float + SSE - 1D FFT Size 512fftw: Float + SSE - 2D FFT Size 64fftw: Stock - 2D FFT Size 64fftw: Float + SSE - 1D FFT Size 32fftw: Float + SSE - 1D FFT Size 128fftw: Float + SSE - 1D FFT Size 64fftw: Stock - 2D FFT Size 256fftw: Float + SSE - 1D FFT Size 256fftw: Stock - 1D FFT Size 512fftw: Stock - 1D FFT Size 32fftw: Float + SSE - 2D FFT Size 32fftw: Float + SSE - 2D FFT Size 128fftw: Stock - 1D FFT Size 256fftw: Stock - 2D FFT Size 128ospray: Magnetic Reconnection - Path Tracerfftw: Stock - 2D FFT Size 32fftw: Stock - 1D FFT Size 128fftw: Stock - 1D FFT Size 64EPYC 76010.04222469.12566.44134233129.2861.554595.224.391.951383.22147.570148762.284.889195065.24.344899.4620.128229.4945272.37621.59156.5220832.2618.6710413.2366.37314.9819.544120.94922088923.114924.12511877.272480.57151.584.08918.123138.11303.9175521.08.7463153.60977.1540.191356.3327742235766316.3280286388.66114.127720233346610.225663261256452.51063616401133416141.8209766481.36306.026308254926420.56329.02008158.46006.75728.3OpenBenchmarking.org

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.7Encoder Mode: Enc Mode 0 - Input: 1080pEPYC 76010.00950.0190.02850.0380.0475SE +/- 0.000, N = 60.0421. (CXX) g++ options: -fPIE -fPIC -pie

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: u8s8f32EPYC 76015K10K15K20K25KSE +/- 59.53, N = 322469.1MIN: 21801.71. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: f32EPYC 76016001200180024003000SE +/- 2.32, N = 32566.44MIN: 2437.381. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096EPYC 76013K6K9K12K15KSE +/- 46.89, N = 3134231. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_all - Data Type: f32EPYC 76017001400210028003500SE +/- 23.55, N = 33129.28MIN: 2970.161. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.5.0Video Input: Chimera 1080p 10-bitEPYC 76011428425670SE +/- 0.15, N = 361.55MIN: 39.53 / MAX: 110.871. (CC) gcc options: -pthread

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 4096EPYC 760110002000300040005000SE +/- 4.25, N = 34595.21. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisEPYC 7601612182430SE +/- 0.00, N = 1224.39MIN: 14.93 / MAX: 26.32

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerEPYC 76010.43880.87761.31641.75522.194SE +/- 0.00, N = 31.95MIN: 1.54 / MAX: 1.98

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f32EPYC 760130060090012001500SE +/- 3.82, N = 31383.22MIN: 1332.561. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32EPYC 7601306090120150SE +/- 1.38, N = 3147.57MIN: 128.591. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 2048EPYC 76013K6K9K12K15KSE +/- 29.63, N = 3148761. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerEPYC 76010.5131.0261.5392.0522.565SE +/- 0.00, N = 32.28MIN: 2.11 / MAX: 2.31

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch 1D - Data Type: f32EPYC 76011.10012.20023.30034.40045.5005SE +/- 0.07959, N = 154.88919MIN: 3.221. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 2048EPYC 760111002200330044005500SE +/- 25.21, N = 35065.21. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisEPYC 76010.97651.9532.92953.9064.8825SE +/- 0.01, N = 34.34MIN: 3.95 / MAX: 4.41

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_3d - Data Type: u8s8f32EPYC 760111002200330044005500SE +/- 3.69, N = 34899.46MIN: 4872.71. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian Dragon ObjEPYC 7601510152025SE +/- 0.05, N = 320.13MIN: 19.92 / MAX: 20.47

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch All - Data Type: f32EPYC 7601714212835SE +/- 0.43, N = 329.49MIN: 27.71. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch All - Data Type: u8s8f32EPYC 760160120180240300SE +/- 0.40, N = 3272.38MIN: 258.781. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: Asian Dragon ObjEPYC 7601510152025SE +/- 0.01, N = 321.59MIN: 21.41 / MAX: 21.92

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: f32EPYC 7601246810SE +/- 0.07721, N = 66.52208MIN: 5.591. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVisEPYC 7601714212835SE +/- 0.00, N = 1232.26MIN: 12.82 / MAX: 33.33

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_3d - Data Type: f32EPYC 7601510152025SE +/- 0.08, N = 318.67MIN: 17.411. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Recurrent Neural Network Training - Data Type: f32EPYC 760190180270360450SE +/- 1.62, N = 3413.24MIN: 395.511. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerEPYC 7601246810SE +/- 0.02, N = 36.37MIN: 5 / MAX: 6.49

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.5.0Video Input: Chimera 1080pEPYC 760170140210280350SE +/- 0.84, N = 3314.98MIN: 247.02 / MAX: 383.841. (CC) gcc options: -pthread

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: CrownEPYC 7601510152025SE +/- 0.04, N = 319.54MIN: 19.21 / MAX: 19.93

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: CrownEPYC 7601510152025SE +/- 0.04, N = 320.95MIN: 20.62 / MAX: 21.36

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 1024EPYC 76014K8K12K16K20KSE +/- 47.22, N = 3208891. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian DragonEPYC 7601612182430SE +/- 0.14, N = 323.11MIN: 22.56 / MAX: 23.56

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: Asian DragonEPYC 7601612182430SE +/- 0.03, N = 324.13MIN: 23.93 / MAX: 24.48

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32EPYC 7601400800120016002000SE +/- 9.10, N = 31877.27MIN: 1855.951. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: u8s8f32EPYC 76015001000150020002500SE +/- 5.12, N = 32480.57MIN: 2430.351. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.5.0Video Input: Summer Nature 4KEPYC 7601306090120150SE +/- 0.45, N = 3151.58MIN: 94.31 / MAX: 164.181. (CC) gcc options: -pthread

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.7Encoder Mode: Enc Mode 4 - Input: 1080pEPYC 76010.921.842.763.684.6SE +/- 0.037, N = 34.0891. (CXX) g++ options: -fPIE -fPIC -pie

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: SciVisEPYC 760148121620SE +/- 0.06, N = 518.12MIN: 9.17 / MAX: 18.52

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32EPYC 76017001400210028003500SE +/- 5.06, N = 33138.11MIN: 3124.281. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: f32EPYC 760170140210280350SE +/- 1.22, N = 3303.92MIN: 291.881. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 1024EPYC 760112002400360048006000SE +/- 6.38, N = 35521.01. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

MKL-DNN DNNL

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

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: f32EPYC 7601246810SE +/- 0.11045, N = 158.74631MIN: 6.951. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch 1D - Data Type: u8s8f32EPYC 76011224364860SE +/- 0.18, N = 353.61MIN: 46.591. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Intel Open Image Denoise

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

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.0.0Scene: MemorialEPYC 7601246810SE +/- 0.03, N = 37.15

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.7Encoder Mode: Enc Mode 8 - Input: 1080pEPYC 7601918273645SE +/- 0.19, N = 340.191. (CXX) g++ options: -fPIE -fPIC -pie

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.5.0Video Input: Summer Nature 1080pEPYC 760180160240320400SE +/- 1.25, N = 3356.33MIN: 203.6 / MAX: 392.891. (CC) gcc options: -pthread

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 4096EPYC 76016K12K18K24K30KSE +/- 42.27, N = 3277421. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 512EPYC 76015K10K15K20K25KSE +/- 25.75, N = 3235761. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 4096EPYC 760114002800420056007000SE +/- 5.98, N = 36316.31. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 2048EPYC 76016K12K18K24K30KSE +/- 284.85, N = 3280281. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 2048EPYC 760114002800420056007000SE +/- 5.48, N = 36388.61. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 512EPYC 760113002600390052006500SE +/- 12.86, N = 36114.11. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 1024EPYC 76016K12K18K24K30KSE +/- 232.71, N = 3277201. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 256EPYC 76015K10K15K20K25KSE +/- 45.35, N = 3233341. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 1024EPYC 760114002800420056007000SE +/- 4.09, N = 36610.21. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 512EPYC 76015K10K15K20K25KSE +/- 85.41, N = 3256631. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 64EPYC 76016K12K18K24K30KSE +/- 32.17, N = 3261251. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 64EPYC 760114002800420056007000SE +/- 6.55, N = 36452.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 32EPYC 76012K4K6K8K10KSE +/- 8.33, N = 3106361. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 128EPYC 76014K8K12K16K20KSE +/- 27.94, N = 3164011. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 64EPYC 76013K6K9K12K15KSE +/- 57.19, N = 3133411. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 256EPYC 760113002600390052006500SE +/- 5.21, N = 36141.81. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 256EPYC 76014K8K12K16K20KSE +/- 137.11, N = 3209761. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 512EPYC 760114002800420056007000SE +/- 4.01, N = 36481.31. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 32EPYC 760114002800420056007000SE +/- 11.62, N = 36306.01. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 32EPYC 76016K12K18K24K30KSE +/- 12.17, N = 3263081. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 128EPYC 76015K10K15K20K25KSE +/- 42.81, N = 3254921. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 256EPYC 760114002800420056007000SE +/- 4.29, N = 36420.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 128EPYC 760114002800420056007000SE +/- 38.57, N = 36329.01. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: Path TracerEPYC 76014080120160200200MIN: 111.11 / MAX: 250

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 32EPYC 76012K4K6K8K10KSE +/- 1.38, N = 38158.41. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 128EPYC 760113002600390052006500SE +/- 3.46, N = 36006.71. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 64EPYC 760112002400360048006000SE +/- 3.36, N = 35728.31. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

72 Results Shown

SVT-AV1
MKL-DNN DNNL:
  Convolution Batch conv_all - u8s8f32
  Convolution Batch conv_all - f32
FFTW
MKL-DNN DNNL
dav1d
FFTW
OSPray:
  San Miguel - SciVis
  San Miguel - Path Tracer
MKL-DNN DNNL:
  Convolution Batch conv_googlenet_v3 - u8s8f32
  Convolution Batch conv_googlenet_v3 - f32
FFTW
OSPray
MKL-DNN DNNL
FFTW
OSPray
MKL-DNN DNNL
Embree
MKL-DNN DNNL:
  IP Batch All - f32
  IP Batch All - u8s8f32
Embree
MKL-DNN DNNL
OSPray
MKL-DNN DNNL:
  Convolution Batch conv_3d - f32
  Recurrent Neural Network Training - f32
OSPray
dav1d
Embree:
  Pathtracer ISPC - Crown
  Pathtracer - Crown
FFTW
Embree:
  Pathtracer ISPC - Asian Dragon
  Pathtracer - Asian Dragon
MKL-DNN DNNL:
  Deconvolution Batch deconv_1d - u8s8f32
  Convolution Batch conv_alexnet - u8s8f32
dav1d
SVT-AV1
OSPray
MKL-DNN DNNL:
  Deconvolution Batch deconv_3d - u8s8f32
  Convolution Batch conv_alexnet - f32
FFTW
MKL-DNN DNNL:
  Deconvolution Batch deconv_3d - f32
  IP Batch 1D - u8s8f32
Intel Open Image Denoise
SVT-AV1
dav1d
FFTW:
  Float + SSE - 1D FFT Size 4096
  Float + SSE - 2D FFT Size 512
  Stock - 1D FFT Size 4096
  Float + SSE - 1D FFT Size 2048
  Stock - 1D FFT Size 2048
  Stock - 2D FFT Size 512
  Float + SSE - 1D FFT Size 1024
  Float + SSE - 2D FFT Size 256
  Stock - 1D FFT Size 1024
  Float + SSE - 1D FFT Size 512
  Float + SSE - 2D FFT Size 64
  Stock - 2D FFT Size 64
  Float + SSE - 1D FFT Size 32
  Float + SSE - 1D FFT Size 128
  Float + SSE - 1D FFT Size 64
  Stock - 2D FFT Size 256
  Float + SSE - 1D FFT Size 256
  Stock - 1D FFT Size 512
  Stock - 1D FFT Size 32
  Float + SSE - 2D FFT Size 32
  Float + SSE - 2D FFT Size 128
  Stock - 1D FFT Size 256
  Stock - 2D FFT Size 128
OSPray
FFTW:
  Stock - 2D FFT Size 32
  Stock - 1D FFT Size 128
  Stock - 1D FFT Size 64

EPYC 7601

Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: ASRockRack EPYCD8 (P2.10 BIOS), Chipset: AMD 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPED1D280GA, Graphics: llvmpipe 126GB, Network: 2 x Intel I350

OS: Ubuntu 19.10, Kernel: 5.4.0-999-generic (x86_64) 20191030, Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 3.3 Mesa 19.2.1 (LLVM 9.0 128 bits), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 1024x768

Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-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 Notes: Scaling Governor: acpi-cpufreq ondemand
Security Notes: 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 STIBP: disabled RSB filling

Testing initiated at 1 November 2019 22:28 by user phoronix.