Talos II Raptor POWER9 System

AMD EPYC vs. Intel Xeon vs. POWER9 CPUs. Benchmarks for a future article.

HTML result view exported from: https://openbenchmarking.org/result/1804042-AR-POWERTALO38.

Talos II Raptor POWER9 SystemProcessorMotherboardMemoryDiskGraphicsAudioNetworkChipsetMonitorOSKernelDisplay DriverCompilerFile-SystemScreen ResolutionDesktopGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551POWER9 altivec supported @ 3.80GHz (64 Cores)PowerNV T2P9D01 REV 1.00262144MB500GB MAXTOR STM350063AMD Radeon Pro WX 7100 8192MBAMD EllesmereBroadcom Limited NetXtreme BCM5719 Gigabit PCIeDebian testing4.16.0-rc4 (ppc64le) 20180307amdgpu 1.4.0GCC 7.3.0ext41024x7682 x Intel Xeon Gold 6138 @ 3.70GHz (40 Cores / 80 Threads)TYAN S7106 (V1.00 BIOS)Intel Sky Lake-E DMI3 Registers12 x 8192 MB DDR4-2666MT/s Micron 9ASF1G72PZ-2G6B1256GB Samsung SSD 850 + 2000GB Seagate ST2000DM006-2DM1 + 2 x 120GB TOSHIBA-TR150ASPEED ASPEED FamilyVE228Intel I210 Gigabit Connection4.16.0-041600-generic (x86_64)GNOME Shell 3.28.01920x1080AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads)TYAN B8026T70AE24HR (V0.05.B10 BIOS)AMD Family 17h8 x 16384 MB DDR4-2666MT/s Samsung M393A2K40BB2-CTD280GB INTEL SSDPE21D280GABroadcom Limited NetXtreme BCM5720 Gigabit PCIeAMD EPYC 7551 32-Core @ 2.00GHz (32 Cores / 64 Threads)GIGABYTE MZ31-AR0-00 v01010101 (F03 BIOS)28672MB2 x Samsung SSD 960 EVO 500GBllvmpipe 28032MBNVIDIA GM204 HD AudioASUS PB278Realtek RTL8111/8168/8411OpenBenchmarking.orgCompiler Details- GCC 7.3.0 Stock Debian Compiler: --build=powerpc64le-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -v - 2 x Intel Xeon Gold 6138: --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++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v - AMD EPYC 7601: --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++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v - AMD EPYC 7551: --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++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v Processor Details- GCC 7.3.0 Stock Debian Compiler: Scaling Governor: powernv-cpufreq ondemand- 2 x Intel Xeon Gold 6138: Scaling Governor: intel_pstate performance- AMD EPYC 7601: Scaling Governor: acpi-cpufreq performance- AMD EPYC 7551: Scaling Governor: acpi-cpufreq performancePython Details- GCC 7.3.0 Stock Debian Compiler: Python 2.7.14+ + Python 3.6.4+- 2 x Intel Xeon Gold 6138: Python 2.7.14+ + Python 3.6.5rc1- AMD EPYC 7601: Python 2.7.14+ + Python 3.6.5rc1- AMD EPYC 7551: Python 2.7.14+ + Python 3.6.5rc1Security Details- 2 x Intel Xeon Gold 6138: KPTI + __user pointer sanitization + Full generic retpoline Protection- AMD EPYC 7601: __user pointer sanitization + Full AMD retpoline Protection- AMD EPYC 7551: __user pointer sanitization + Full AMD retpoline Protection

Talos II Raptor POWER9 Systemparboil: OpenMP LBMparboil: OpenMP CUTCPparboil: OpenMP Stencilparboil: OpenMP MRI Griddingrodinia: OpenMP LavaMDrodinia: OpenMP CFD Solverrodinia: OpenMP Streamclusterx264: H.264 Video Encodingcompress-7zip: Compress Speed Testbuild-gcc: Time To Compilebuild-linux-kernel: Time To Compilec-ray: Total Timestockfish: Total Timeencode-flac: WAV To FLACencode-mp3: WAV To MP3openssl: RSA 4096-bit Performanceredis: GETredis: SETpybench: Total For Average Test Timesphpbench: PHP Benchmark Suiteosbench: Create Threadsosbench: Create Processesosbench: Memory AllocationsGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 755166.089.6910.51817.4438.5819.0927.8043.72824471070.7072.614.65491551.7975.272104.871049994.41606874.10485916320827.1729.7783.0330.182.286.01397.4327.1010.0014.51127.28143505591.3227.803.15334310.2733.207965.402515784.921744256.79139560246723.0742.9596.0537.392.6114.26256.1532.5910.7117.46128.1799574707.3435.663.46447411.7943.574598.471703353.371195935.33208639365930.7159.6195.1471.882.7617.35355.6334.7912.2728.79101.5279708926.0839.303.59503212.7145.694387.471639124.131161953.33221636576738.2557.9596.32OpenBenchmarking.org

Parboil

Test: OpenMP LBM

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP LBMGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75511632486480SE +/- 0.20, N = 3SE +/- 0.75, N = 6SE +/- 0.38, N = 3SE +/- 1.94, N = 666.0830.1837.3971.881. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenMP CUTCP

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP CUTCPGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75513691215SE +/- 0.14, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 39.692.282.612.761. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenMP Stencil

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP StencilGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 755148121620SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.88, N = 6SE +/- 0.49, N = 610.516.0114.2617.351. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenMP MRI Gridding

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP MRI GriddingGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75512004006008001000SE +/- 0.64, N = 3SE +/- 8.35, N = 6SE +/- 2.31, N = 3SE +/- 1.42, N = 3817.44397.43256.15355.631. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LavaMDGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551918273645SE +/- 0.05, N = 3SE +/- 0.22, N = 3SE +/- 0.14, N = 3SE +/- 0.29, N = 338.5827.1032.5934.791. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP CFD Solver

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP CFD SolverGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551510152025SE +/- 0.16, N = 3SE +/- 0.20, N = 6SE +/- 0.14, N = 3SE +/- 0.31, N = 619.0910.0010.7112.271. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP StreamclusterGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551714212835SE +/- 0.08, N = 3SE +/- 0.50, N = 6SE +/- 0.53, N = 6SE +/- 3.79, N = 627.8014.5117.4628.791. (CXX) g++ options: -O2 -lOpenCL

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2018-02-05H.264 Video EncodingGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551306090120150SE +/- 0.14, N = 3SE +/- 5.54, N = 6SE +/- 0.67, N = 3SE +/- 1.55, N = 343.72127.28128.17101.52-maltivec -mabi=altivec -mvsx-m64-m64-m641. (CC) gcc options: -ldl -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

7-Zip Compression

Compress Speed Test

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 16.02Compress Speed TestGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 755130K60K90K120K150KSE +/- 390.52, N = 3SE +/- 2372.97, N = 3SE +/- 669.88, N = 3SE +/- 2630.45, N = 68244714350599574797081. (CXX) g++ options: -pipe -lpthread

Timed GCC Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GCC Compilation 7.2Time To CompileGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75512004006008001000SE +/- 2.17, N = 3SE +/- 0.65, N = 3SE +/- 5.02, N = 3SE +/- 5.33, N = 31070.70591.32707.34926.08

Timed Linux Kernel Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 4.13Time To CompileGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75511632486480SE +/- 1.20, N = 4SE +/- 0.40, N = 5SE +/- 0.47, N = 6SE +/- 0.75, N = 672.6127.8035.6639.30

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75511.04632.09263.13894.18525.2315SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 34.653.153.463.591. (CC) gcc options: -lm -lpthread -O3

Stockfish

Total Time

OpenBenchmarking.orgms, Fewer Is BetterStockfish 2014-11-26Total TimeGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 755111002200330044005500SE +/- 1.33, N = 3SE +/- 10.04, N = 3SE +/- 167.95, N = 64915334344745032-msse -msse3 -mpopcnt-msse -msse3 -mpopcnt-msse -msse3 -mpopcnt1. (CXX) g++ options: -lpthread -fno-exceptions -fno-rtti -ansi -pedantic -O3 -flto

FLAC Audio Encoding

WAV To FLAC

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.2WAV To FLACGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75511224364860SE +/- 0.25, N = 5SE +/- 0.02, N = 5SE +/- 0.01, N = 5SE +/- 0.09, N = 551.7910.2711.7912.71-logg1. (CXX) g++ options: -O2 -fvisibility=hidden -lm

LAME MP3 Encoding

WAV To MP3

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3GCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 755120406080100SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.52, N = 3SE +/- 0.07, N = 375.2733.2043.5745.69-lncurses1. (CC) gcc options: -lm

OpenSSL

RSA 4096-bit Performance

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.0fRSA 4096-bit PerformanceGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75512K4K6K8K10KSE +/- 1.27, N = 3SE +/- 50.05, N = 3SE +/- 22.34, N = 3SE +/- 0.85, N = 32104.877965.404598.474387.471. (CC) gcc options: -O3 -pthread -m64 -lssl -lcrypto -ldl

Redis

Test: GET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: GETGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551500K1000K1500K2000K2500KSE +/- 12692.76, N = 3SE +/- 108856.52, N = 6SE +/- 25113.28, N = 3SE +/- 41782.46, N = 61049994.412515784.921703353.371639124.131. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: SET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: SETGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551400K800K1200K1600K2000KSE +/- 11534.54, N = 3SE +/- 62786.06, N = 6SE +/- 23759.98, N = 6SE +/- 6077.00, N = 3606874.101744256.791195935.331161953.331. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

PyBench

Total For Average Test Times

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 755110002000300040005000SE +/- 4.84, N = 3SE +/- 2.60, N = 3SE +/- 1.45, N = 3SE +/- 3.61, N = 34859139520862216

PHPBench

PHP Benchmark Suite

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551130K260K390K520K650KSE +/- 55.73, N = 3SE +/- 1392.88, N = 3SE +/- 604.68, N = 3SE +/- 2983.18, N = 3163208602467393659365767

OSBench

Test: Create Threads

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ThreadsGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7551918273645SE +/- 0.45, N = 4SE +/- 0.28, N = 3SE +/- 0.51, N = 3SE +/- 0.61, N = 627.1723.0730.7138.25-lm-lm-lm1. (CC) gcc options:

OSBench

Test: Create Processes

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ProcessesGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 75511326395265SE +/- 0.19, N = 3SE +/- 0.89, N = 6SE +/- 0.87, N = 4SE +/- 0.23, N = 329.7742.9559.6157.95-lm-lm-lm1. (CC) gcc options:

OSBench

Test: Memory Allocations

OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory AllocationsGCC 7.3.0 Stock Debian Compiler2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 755120406080100SE +/- 0.67, N = 3SE +/- 0.42, N = 3SE +/- 0.05, N = 3SE +/- 1.87, N = 683.0396.0595.1496.32-lm-lm-lm1. (CC) gcc options:

Phoronix Test Suite v10.8.4