AMD EPYC vs. Xeon - Ubuntu Linux

Some initial AMD EPYC 7601 tests on Ubuntu 17.04 with Linux 4.13. Tests for a future article on Phoronix.com. Benchmarks by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/1709122-TY-EPYC1196225&rdt&grs.

AMD EPYC vs. Xeon - Ubuntu LinuxProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkAudioOSKernelDisplay DriverCompilerFile-SystemScreen ResolutionDesktopAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v3AMD EPYC 7601 32-Core @ 2.20GHz (64 Cores)TYAN B8026T70AE24HRAMD Device 1450129024MB234GBASPEED ASPEED FamilyAcer P243WBroadcom Limited NetXtreme BCM5720 Gigabit PCIeUbuntu 17.044.13.0-041300-generic (x86_64)modesetting 1.19.3GCC 6.3.0 20170406ext41920x1200Unity 7.5.02 x Intel Xeon Gold 6138 @ 3.70GHz (80 Cores)TYAN S7106Intel Device 202096256MB256GB Samsung SSD 850 + 2000GB Seagate ST2000DM006-2DM1 + 2 x 120GB TOSHIBA-TR150Intel I210 Gigabit ConnectionIntel Xeon E5-2687W v3 @ 3.50GHz (20 Cores)MSI X99S SLI PLUS (MS-7885) v1.0Intel Xeon E7 v3/Xeon32768MB80GB INTEL SSDSCKGW08NVIDIA GeForce GTX 770 2048MBRealtek ALC892DELL S2409WIntel Connection1920x1080Intel Xeon E5-1680 v3 @ 3.80GHz (16 Cores)ASUS X99-A16384MB60GB Patriot TorcheVGA NVIDIA GeForce GT 740 1024MBRealtek ALC1150OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-browser-plugin --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --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=auto --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-arch-directory=amd64 --with-default-libstdcxx-abi=new --with-multilib-list=m32,m64,mx32 --with-target-system-zlib --with-tune=generic -v Processor Details- AMD EPYC 7601: Scaling Governor: acpi-cpufreq ondemand- AMD EPYC 7601 (NUMA Interleave All): Scaling Governor: acpi-cpufreq ondemand- 2 x Intel Xeon Gold 6138: Scaling Governor: intel_pstate powersave- Intel Xeon E5-2687W v3: Scaling Governor: intel_pstate powersave- Intel Xeon E5-1680 v3: Scaling Governor: intel_pstate powersave

AMD EPYC vs. Xeon - Ubuntu Linuxc-ray: Total Timeopenssl: RSA 4096-bit Performancenpb: LU.Crodinia: OpenMP LavaMDprimesieve: 1e12 Prime Number Generationbuild-llvm: Time To Compileparboil: OpenMP LBMnpb: EP.Cdarktable: Masskrug - CPU-onlyblender: BMW27 - CPU-Onlyx264: H.264 Video Encodingdarktable: Server Room - CPU-onlydarktable: Boat - CPU-onlybuild-linux-kernel: Time To Compilettsiod-renderer: Phong Rendering With Soft-Shadow Mappingjohn-the-ripper: Blowfishrodinia: OpenMP Streamclusterparboil: OpenMP Stencilnpb: LU.AAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v32.843294.5346983.3430.9614.08175.1150.951610.179.24621.29292.634.627.9437.40412.712955323.1513.8263042.802.853306.3750046.9030.1414.06192.2638.371607.687.57625.93288.283.727.0339.28430.153433514.667.7562479.082.844826.7050072.2531.4411.81134.6450.041815.8913.14540.98310.795.2914.3030.54236.463037322.527.7553974.4910.531339.0016108.0393.9336.33388.12108.07686.328.34556.58311.337.0114.0273.02362.431452027.2912.6119699.8611.091271.2714551.9598.3838.20408.0790.01660.248.39475.88262.136.9712.9275.83355.971381526.2012.7420916.24OpenBenchmarking.org

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v33691215SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 32.842.852.8410.5311.091. (CC) gcc options: -lm -lpthread -O3

OpenSSL

RSA 4096-bit Performance

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.0.1gRSA 4096-bit PerformanceAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v310002000300040005000SE +/- 14.45, N = 3SE +/- 13.22, N = 3SE +/- 23.22, N = 3SE +/- 0.46, N = 3SE +/- 0.38, N = 33294.533306.374826.701339.001271.271. (CC) gcc options: -m64 -O3 -lssl -lcrypto -ldl

NAS Parallel Benchmarks

Test / Class: LU.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: LU.CAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v311K22K33K44K55KSE +/- 747.98, N = 3SE +/- 25.06, N = 3SE +/- 637.78, N = 3SE +/- 89.16, N = 3SE +/- 55.08, N = 346983.3450046.9050072.2516108.0314551.951. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 2.0.2

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LavaMDAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v320406080100SE +/- 0.01, N = 3SE +/- 0.15, N = 3SE +/- 0.10, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 330.9630.1431.4493.9398.381. (CXX) g++ options: -O2 -lOpenCL

Primesieve

1e12 Prime Number Generation

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 5.4.21e12 Prime Number GenerationAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v3918273645SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 314.0814.0611.8136.3338.201. (CXX) g++ options: -O2 -fopenmp

Timed LLVM Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 4.0.1Time To CompileAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v390180270360450SE +/- 3.02, N = 4SE +/- 2.34, N = 3SE +/- 0.81, N = 3SE +/- 0.11, N = 3SE +/- 0.61, N = 3175.11192.26134.64388.12408.07

Parboil

Test: OpenMP LBM

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP LBMAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v320406080100SE +/- 0.66, N = 3SE +/- 0.26, N = 3SE +/- 1.14, N = 6SE +/- 0.10, N = 3SE +/- 0.02, N = 350.9538.3750.04108.0790.011. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

NAS Parallel Benchmarks

Test / Class: EP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: EP.CAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v3400800120016002000SE +/- 0.32, N = 3SE +/- 0.90, N = 3SE +/- 34.76, N = 6SE +/- 3.03, N = 3SE +/- 0.54, N = 31610.171607.681815.89686.32660.241. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 2.0.2

Darktable

Test: Masskrug - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.2.1Test: Masskrug - Acceleration: CPU-onlyAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v33691215SE +/- 0.09, N = 3SE +/- 0.13, N = 4SE +/- 0.27, N = 6SE +/- 0.04, N = 3SE +/- 0.03, N = 39.247.5713.148.348.39

Blender

Blend File: BMW27 - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.78cBlend File: BMW27 - Compute: CPU-OnlyAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v3140280420560700621.29625.93540.98556.58475.88

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2017-09-08H.264 Video EncodingAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v370140210280350SE +/- 1.27, N = 3SE +/- 0.45, N = 3SE +/- 3.63, N = 3SE +/- 1.06, N = 3SE +/- 2.72, N = 3292.63288.28310.79311.33262.13-lavformat -lavcodec -lavutil -lswscale-lavformat -lavcodec -lavutil -lswscale1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

Darktable

Test: Server Room - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.2.1Test: Server Room - Acceleration: CPU-onlyAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v3246810SE +/- 0.07, N = 6SE +/- 0.08, N = 6SE +/- 0.35, N = 6SE +/- 0.05, N = 3SE +/- 0.00, N = 34.623.725.297.016.97

Darktable

Test: Boat - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.2.1Test: Boat - Acceleration: CPU-onlyAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v348121620SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.59, N = 6SE +/- 0.01, N = 3SE +/- 0.02, N = 37.947.0314.3014.0212.92

Timed Linux Kernel Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 4.9Time To CompileAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v320406080100SE +/- 0.51, N = 6SE +/- 0.58, N = 5SE +/- 0.89, N = 6SE +/- 0.75, N = 3SE +/- 1.05, N = 637.4039.2830.5473.0275.83

TTSIOD 3D Renderer

Phong Rendering With Soft-Shadow Mapping

OpenBenchmarking.orgFPS, More Is BetterTTSIOD 3D Renderer 2.3aPhong Rendering With Soft-Shadow MappingAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v390180270360450SE +/- 4.72, N = 3SE +/- 3.57, N = 3SE +/- 17.30, N = 6SE +/- 0.54, N = 3SE +/- 0.43, N = 3412.71430.15236.46362.43355.971. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -lstdc++

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.8.0Test: BlowfishAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v37K14K21K28K35KSE +/- 2253.07, N = 6SE +/- 764.17, N = 6SE +/- 2076.33, N = 6SE +/- 9.33, N = 329553343353037314520138151. (CC) gcc options: -fopenmp -lcrypt

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP StreamclusterAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v3612182430SE +/- 1.43, N = 6SE +/- 0.27, N = 6SE +/- 0.42, N = 3SE +/- 0.02, N = 3SE +/- 0.59, N = 623.1514.6622.5227.2926.201. (CXX) g++ options: -O2 -lOpenCL

Parboil

Test: OpenMP Stencil

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP StencilAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v348121620SE +/- 0.68, N = 6SE +/- 0.00, N = 3SE +/- 0.17, N = 6SE +/- 0.05, N = 3SE +/- 0.01, N = 313.827.757.7512.6112.741. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

NAS Parallel Benchmarks

Test / Class: LU.A

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: LU.AAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138Intel Xeon E5-2687W v3Intel Xeon E5-1680 v314K28K42K56K70KSE +/- 1051.01, N = 4SE +/- 437.89, N = 3SE +/- 3783.72, N = 6SE +/- 19.47, N = 3SE +/- 11.47, N = 363042.8062479.0853974.4919699.8620916.241. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 2.0.2

Phoronix Test Suite v10.8.5