AMD EPYC Ubuntu 2017 vs. 2018 Linux Benchmarks

AMD EPYC 2018 Linux benchmarking by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/1806110-AR-AMDEPYC2481&sor.

ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelCompilerFile-SystemScreen Resolution20172018 EPYC 7351P EPYC 7401P EPYC 7601 EPYC 7351P EPYC 7401P EPYC 7601AMD EPYC 7351P 16-Core @ 2.40GHz (16 Cores / 32 Threads)TYAN B8026T70AE24HR (V1.02.B10 BIOS)AMD Device 1450129024MB280GB INTEL SSDPE21D280GAASPEED ASPEED FamilyVE228Broadcom NetXtreme BCM5720 Gigabit PCIeUbuntu 16.044.4.0-127-generic (x86_64)GCC 5.4.0 20160609ext41920x1080AMD EPYC 7401P 24-Core @ 2.00GHz (24 Cores / 48 Threads)AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads)AMD EPYC 7351P 16-Core @ 2.40GHz (16 Cores / 32 Threads)AMD Family 17hBroadcom Limited NetXtreme BCM5720 Gigabit PCIeUbuntu 18.104.17.0-041700-generic (x86_64)GCC 8.1.0AMD EPYC 7401P 24-Core @ 2.00GHz (24 Cores / 48 Threads)AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads)OpenBenchmarking.orgEnvironment Details- CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Details- 2017: EPYC 7351P: --build=x86_64-linux-gnu --disable-browser-plugin --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-arch-directory=amd64 --with-default-libstdcxx-abi=new --with-multilib-list=m32,m64,mx32 --with-tune=generic -v - 2017: EPYC 7401P: --build=x86_64-linux-gnu --disable-browser-plugin --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-arch-directory=amd64 --with-default-libstdcxx-abi=new --with-multilib-list=m32,m64,mx32 --with-tune=generic -v - 2017: EPYC 7601: --build=x86_64-linux-gnu --disable-browser-plugin --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-arch-directory=amd64 --with-default-libstdcxx-abi=new --with-multilib-list=m32,m64,mx32 --with-tune=generic -v - 2018: EPYC 7351P: --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 - 2018: EPYC 7401P: --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 - 2018: 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 Processor Details- Scaling Governor: acpi-cpufreq ondemandPython Details- 2017: EPYC 7351P: Python 2.7.12 + Python 3.5.2- 2017: EPYC 7401P: Python 2.7.12 + Python 3.5.2- 2017: EPYC 7601: Python 2.7.12 + Python 3.5.2- 2018: EPYC 7351P: Python 2.7.15 + Python 3.6.5- 2018: EPYC 7401P: Python 2.7.15 + Python 3.6.5- 2018: EPYC 7601: Python 2.7.15 + Python 3.6.5Security Details- 2017: EPYC 7351P: OSB (observable speculation barrier Intel v6) + Full AMD retpoline IBPB (Intel v4) + SSB disabled via prctl and seccomp Protection- 2017: EPYC 7401P: OSB (observable speculation barrier Intel v6) + Full AMD retpoline IBPB (Intel v4) + SSB disabled via prctl and seccomp Protection- 2017: EPYC 7601: OSB (observable speculation barrier Intel v6) + Full AMD retpoline IBPB (Intel v4) + SSB disabled via prctl and seccomp Protection- 2018: EPYC 7351P: __user pointer sanitization + Full AMD retpoline IBPB + SSB disabled via prctl and seccomp Protection- 2018: EPYC 7401P: __user pointer sanitization + Full AMD retpoline IBPB + SSB disabled via prctl and seccomp Protection- 2018: EPYC 7601: __user pointer sanitization + Full AMD retpoline IBPB + SSB disabled via prctl and seccomp Protection

npb: EP.Cparboil: OpenMP LBMparboil: OpenMP CUTCPparboil: OpenMP Stencilparboil: OpenMP MRI Griddinghpcg: dolfyn: Computational Fluid Dynamicshmmer: Pfam Database Searchjava-gradle-perf: Reactorjohn-the-ripper: Blowfishjohn-the-ripper: Traditional DESnero2d: Total Timettsiod-renderer: Phong Rendering With Soft-Shadow Mappingx264: H.264 Video Encodingcompress-7zip: Compress Speed Teststockfish: Total Timeasmfish: 1024 Hash Memory, 26 Depthebizzy: build-gcc: Time To Compilebuild-llvm: Time To Compilec-ray: Total Timecompress-pbzip2: 256MB File Compressionpovray: Trace Timeprimesieve: 1e12 Prime Number Generationm-queens: Time To Solveopenssl: RSA 4096-bit Performancetensorflow: Cifar10darktable: Boat - CPU-onlydarktable: Masskrug - CPU-onlydarktable: Server Room - CPU-onlyblender: Barbershop - CPU-Onlypybench: Total For Average Test Timesapache: Static Web Page Servingpgbench: Buffer Test - Normal Load - Read Writepgbench: Buffer Test - Normal Load - Read Only20172018 EPYC 7351P EPYC 7401P EPYC 7601 EPYC 7351P EPYC 7401P EPYC 7601890.3250.744.3713.85135.530.6224.508.5827.40199078132533338.76332.4394.116291033618684379748556151991033.01238.224.762.9436.3719.9954.732494.9064.026.560.280.28850.84223918040.1626185.34216093.171243.3841.723.6110.69222.730.6123.638.3627.952802710560733333.26358.29109.38824934301940148848106686968982.50199.923.432.5828.5716.1742.043370.8058.946.390.290.27668.54216617800.2328709.51279449.041594.1934.562.817.95298.620.6822.248.0928.873637213828466729.77419.93113.131015145901149866722944687212935.09162.822.732.2522.0112.3032.484561.7756.006.330.270.27504.40202917289.9631743.29351648.77946.0457.834.0112.46142.950.748.4625.39357.36641003233375037112247840031985.614.642.9220.9553.652497.5765.85852.56179915814.6928338.14208838.541312.0643.743.349.60226.310.748.4525.81408.588451042461968488856761034390945.793.332.4216.8141.503379.4760.88666.80172815205.7934345.77268314.911687.9634.892.587.77319.450.758.4726.37458.7210249257376077661081781175371898.792.572.1512.7732.034571.0358.45505.41162815722.2435922.09402237.65OpenBenchmarking.org

NAS Parallel Benchmarks

Test / Class: EP.C

20182017OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: EP.CEPYC 7601EPYC 7401PEPYC 7351P400800120016002000SE +/- 3.28, N = 3SE +/- 10.87, N = 3SE +/- 2.38, N = 3SE +/- 1.24, N = 3SE +/- 5.32, N = 3SE +/- 0.83, N = 31687.961312.06946.041594.191243.38890.321. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 1.10.2

Parboil

Test: OpenMP LBM

20172018OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP LBMEPYC 7601EPYC 7401PEPYC 7351P1326395265SE +/- 0.20, N = 3SE +/- 0.56, N = 3SE +/- 0.25, N = 3SE +/- 0.44, N = 3SE +/- 0.96, N = 6SE +/- 0.48, N = 334.5641.7250.7434.8943.7457.831. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenMP CUTCP

20182017OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP CUTCPEPYC 7601EPYC 7401PEPYC 7351P0.98331.96662.94993.93324.9165SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 32.583.344.012.813.614.371. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenMP Stencil

20182017OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP StencilEPYC 7601EPYC 7401PEPYC 7351P48121620SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.19, N = 6SE +/- 0.19, N = 37.779.6012.467.9510.6913.851. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenMP MRI Gridding

20172018OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP MRI GriddingEPYC 7351PEPYC 7401PEPYC 760170140210280350SE +/- 0.61, N = 3SE +/- 1.10, N = 3SE +/- 0.82, N = 3SE +/- 0.45, N = 3SE +/- 0.58, N = 3SE +/- 0.86, N = 3135.53222.73298.62142.95226.31319.451. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

High Performance Conjugate Gradient

20182017OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.0EPYC 7601EPYC 7401PEPYC 7351P0.16880.33760.50640.67520.844SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 30.750.740.740.680.620.61

Dolfyn

Computational Fluid Dynamics

OpenBenchmarking.orgSeconds, Fewer Is BetterDolfyn 0.527Computational Fluid DynamicsEPYC 7601EPYC 7401PEPYC 7351P612182430SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 322.2423.6324.50

Timed HMMer Search

Pfam Database Search

20172018OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 2.3.2Pfam Database SearchEPYC 7601EPYC 7401PEPYC 7351P246810SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 38.098.368.588.458.468.471. (CC) gcc options: -O3 -march=native -pthread -lhmmer -lsquid -lm

Java Gradle Build

Gradle Build: Reactor

20182017OpenBenchmarking.orgSeconds, Fewer Is BetterJava Gradle Build 1.0Gradle Build: ReactorEPYC 7351PEPYC 7401PEPYC 7601714212835SE +/- 0.07, N = 4SE +/- 0.12, N = 4SE +/- 0.03, N = 4SE +/- 0.12, N = 4SE +/- 0.12, N = 4SE +/- 0.04, N = 425.3925.8126.3727.4027.9528.87

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.8.0-jumbo-1Test: BlowfishEPYC 7601EPYC 7401PEPYC 7351P8K16K24K32K40KSE +/- 221.23, N = 3SE +/- 246.54, N = 3SE +/- 15.86, N = 33637228027199071. (CC) gcc options: -lssl -lcrypto -fopenmp -pthread -lm -lz -ldl -lcrypt -lbz2

John The Ripper

Test: Traditional DES

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.8.0-jumbo-1Test: Traditional DESEPYC 7601EPYC 7401PEPYC 7351P30M60M90M120M150MSE +/- 1755962.07, N = 3SE +/- 1919243.72, N = 3SE +/- 376251.84, N = 3138284667105607333813253331. (CC) gcc options: -lssl -lcrypto -fopenmp -pthread -lm -lz -ldl -lcrypt -lbz2

Open FMM Nero2D

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen FMM Nero2D 2.0.2Total TimeEPYC 7601EPYC 7401PEPYC 7351P918273645SE +/- 0.21, N = 3SE +/- 0.41, N = 3SE +/- 0.33, N = 329.7733.2638.761. (CXX) g++ options: -O3 -march=native -lfftw3 -llapack -lf77blas -latlas -lgfortran -lquadmath -lm -pthread -lmpi_cxx -lmpi

TTSIOD 3D Renderer

Phong Rendering With Soft-Shadow Mapping

20182017OpenBenchmarking.orgFPS, More Is BetterTTSIOD 3D Renderer 2.3aPhong Rendering With Soft-Shadow MappingEPYC 7601EPYC 7401PEPYC 7351P100200300400500SE +/- 0.25, N = 3SE +/- 0.49, N = 3SE +/- 0.53, N = 3SE +/- 1.31, N = 3SE +/- 1.35, N = 3SE +/- 0.39, N = 3458.72408.58357.36419.93358.29332.431. (CXX) g++ options: -O3 -march=native -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -lstdc++

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2018-02-05H.264 Video EncodingEPYC 7601EPYC 7401PEPYC 7351P306090120150SE +/- 0.89, N = 3SE +/- 0.69, N = 3SE +/- 0.33, N = 3113.13109.3894.111. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -march=native -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

7-Zip Compression

Compress Speed Test

20182017OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 16.02Compress Speed TestEPYC 7601EPYC 7401PEPYC 7351P20K40K60K80K100KSE +/- 289.75, N = 3SE +/- 298.68, N = 3SE +/- 180.74, N = 3SE +/- 456.16, N = 3SE +/- 335.22, N = 3SE +/- 79.60, N = 3102492845106410010151482493629101. (CXX) g++ options: -pipe -lpthread

Stockfish

Total Time

20172018OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 9Total TimeEPYC 7601EPYC 7401PEPYC 7351P13M26M39M52M65MSE +/- 67441.67, N = 3SE +/- 248095.00, N = 3SE +/- 247766.79, N = 3SE +/- 304021.78, N = 3SE +/- 332946.60, N = 3SE +/- 111771.25, N = 35901149843019401336186845737607742461968323337501. (CXX) g++ options: -m64 -lpthread -O3 -march=native -fno-exceptions -std=c++11 -pedantic -msse -msse3 -mpopcnt -flto

asmFish

1024 Hash Memory, 26 Depth

20172018OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2017-09-191024 Hash Memory, 26 DepthEPYC 7601EPYC 7401PEPYC 7351P14M28M42M56M70MSE +/- 794446.07, N = 3SE +/- 275602.30, N = 3SE +/- 514739.98, N = 3SE +/- 404906.57, N = 3SE +/- 540357.31, N = 3SE +/- 275491.98, N = 3667229444884810637974855661081784888567637112247

ebizzy

20182017OpenBenchmarking.orgRecords/s, More Is Betterebizzy 0.3EPYC 7601EPYC 7401PEPYC 7351P300K600K900K1200K1500KSE +/- 14977.22, N = 3SE +/- 17943.77, N = 3SE +/- 14624.57, N = 3SE +/- 19526.02, N = 6SE +/- 12101.52, N = 3SE +/- 10696.32, N = 4117537110343908400316872126869686151991. (CC) gcc options: -pthread -lpthread -O3 -march=native

Timed GCC Compilation

Time To Compile

20182017OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GCC Compilation 7.2Time To CompileEPYC 7601EPYC 7401PEPYC 7351P2004006008001000SE +/- 0.35, N = 3SE +/- 0.67, N = 3SE +/- 0.13, N = 3SE +/- 0.74, N = 3SE +/- 0.43, N = 3SE +/- 0.70, N = 3898.79945.79985.61935.09982.501033.01

Timed LLVM Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 4.0.1Time To CompileEPYC 7601EPYC 7401PEPYC 7351P50100150200250SE +/- 1.09, N = 3SE +/- 0.32, N = 3SE +/- 2.53, N = 3162.82199.92238.22

C-Ray

Total Time

20182017OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeEPYC 7601EPYC 7401PEPYC 7351P1.0712.1423.2134.2845.355SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 32.573.334.642.733.434.761. (CC) gcc options: -lm -lpthread -O3 -march=native

Parallel BZIP2 Compression

256MB File Compression

20182017OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.12256MB File CompressionEPYC 7601EPYC 7401PEPYC 7351P0.66151.3231.98452.6463.3075SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 32.152.422.922.252.582.941. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

POV-Ray

Trace Time

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace TimeEPYC 7601EPYC 7401PEPYC 7351P816243240SE +/- 0.21, N = 3SE +/- 0.19, N = 3SE +/- 0.08, N = 322.0128.5736.371. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lXpm -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

Primesieve

1e12 Prime Number Generation

20172018OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 6.21e12 Prime Number GenerationEPYC 7601EPYC 7401PEPYC 7351P510152025SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 312.3016.1719.9912.7716.8120.951. (CXX) g++ options: -O3 -march=native -rdynamic -lpthread

m-queens

Time To Solve

20182017OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.1Time To SolveEPYC 7601EPYC 7401PEPYC 7351P1224364860SE +/- 0.12, N = 3SE +/- 0.08, N = 3SE +/- 0.13, N = 3SE +/- 0.14, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 332.0341.5053.6532.4842.0454.731. (CXX) g++ options: -fopenmp -O3 -march=native -O2

OpenSSL

RSA 4096-bit Performance

20182017OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.0fRSA 4096-bit PerformanceEPYC 7601EPYC 7401PEPYC 7351P10002000300040005000SE +/- 6.55, N = 3SE +/- 10.40, N = 3SE +/- 2.55, N = 3SE +/- 10.77, N = 3SE +/- 10.59, N = 3SE +/- 0.93, N = 34571.033379.472497.574561.773370.802494.901. (CC) gcc options: -O3 -pthread -m64 -lssl -lcrypto -ldl

Tensorflow

Build: Cifar10

20172018OpenBenchmarking.orgSeconds, Fewer Is BetterTensorflow 2017-02-03Build: Cifar10EPYC 7601EPYC 7401PEPYC 7351P1530456075SE +/- 0.09, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 356.0058.9464.0258.4560.8865.85

Darktable

Test: Boat - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.0.3Test: Boat - Acceleration: CPU-onlyEPYC 7601EPYC 7401PEPYC 7351P246810SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 36.336.396.56

Darktable

Test: Masskrug - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.0.3Test: Masskrug - Acceleration: CPU-onlyEPYC 7601EPYC 7351PEPYC 7401P0.06530.13060.19590.26120.3265SE +/- 0.00, N = 3SE +/- 0.01, N = 6SE +/- 0.02, N = 60.270.280.29

Darktable

Test: Server Room - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.0.3Test: Server Room - Acceleration: CPU-onlyEPYC 7401PEPYC 7601EPYC 7351P0.0630.1260.1890.2520.315SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.270.270.28

Blender

Blend File: Barbershop - Compute: CPU-Only

20172018OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.79aBlend File: Barbershop - Compute: CPU-OnlyEPYC 7601EPYC 7401PEPYC 7351P2004006008001000504.40668.54850.84505.41666.80852.56

PyBench

Total For Average Test Times

20182017OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesEPYC 7601EPYC 7401PEPYC 7351P5001000150020002500SE +/- 3.84, N = 3SE +/- 3.71, N = 3SE +/- 6.77, N = 3SE +/- 2.91, N = 3SE +/- 4.98, N = 3SE +/- 2.65, N = 3162817281799202921662239

Apache Benchmark

Static Web Page Serving

20172018OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.29Static Web Page ServingEPYC 7351PEPYC 7401PEPYC 76014K8K12K16K20KSE +/- 72.33, N = 3SE +/- 352.97, N = 3SE +/- 138.05, N = 3SE +/- 15.24, N = 3SE +/- 250.60, N = 6SE +/- 10.09, N = 318040.1617800.2317289.9615814.6915722.2415205.791. (CC) gcc options: -shared -fPIC -pthread -O3 -march=native

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Write

20182017OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.3Scaling: Buffer Test - Test: Normal Load - Mode: Read WriteEPYC 7601EPYC 7401PEPYC 7351P8K16K24K32K40KSE +/- 100.74, N = 3SE +/- 64.52, N = 3SE +/- 41.93, N = 3SE +/- 218.49, N = 3SE +/- 130.88, N = 3SE +/- 139.16, N = 335922.0934345.7728338.1431743.2928709.5126185.341. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Only

20182017OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.3Scaling: Buffer Test - Test: Normal Load - Mode: Read OnlyEPYC 7601EPYC 7401PEPYC 7351P90K180K270K360K450KSE +/- 3888.22, N = 3SE +/- 1970.14, N = 3SE +/- 1108.72, N = 3SE +/- 5929.59, N = 6SE +/- 4630.97, N = 6SE +/- 527.88, N = 3402237.65268314.91208838.54351648.77279449.04216093.171. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

Phoronix Test Suite v10.8.4