AMD EPYC vs. Intel Xeon vs. Amazon EC2 Instances

Tests for a future article. 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/1709185-TY-EC2COMPUT15&grr&sor.

AMD EPYC vs. Intel Xeon vs. Amazon EC2 InstancesProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDisplay DriverCompilerFile-SystemScreen ResolutionDesktopSystem LayerAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138m4.10xlargem4.16xlargec4.4xlargec4.8xlargeAMD 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 Connection2 x Intel Xeon E5-2676 v3 @ 3.00GHz (40 Cores)Xen HVM domUIntel 440FX- 82441FX PMC161792MB8GBCirrus Logic GD 5446Intel 82599 Virtual FunctionUbuntu 16.044.4.0-1022-aws (x86_64)GCC 5.4.0 20160609Xen HVM domU 4.2.amazon2 x Intel Xeon E5-2686 v4 @ 3.00GHz (64 Cores)258048MBDevice 1d0f:ec20Intel Xeon E5-2666 v3 @ 2.90GHz (16 Cores)30720MBIntel 82599 Virtual Function2 x Intel Xeon E5-2666 v3 @ 3.50GHz (36 Cores)60416MBOpenBenchmarking.orgCompiler Details- AMD EPYC 7601: --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 - AMD EPYC 7601 (NUMA Interleave All): --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 - 2 x Intel Xeon Gold 6138: --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 - m4.10xlarge: --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 - m4.16xlarge: --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 - c4.4xlarge: --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 - c4.8xlarge: --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 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- m4.10xlarge: Scaling Governor: intel_pstate powersave- m4.16xlarge: Scaling Governor: intel_pstate powersave- c4.8xlarge: Scaling Governor: intel_pstate powersave

AMD EPYC vs. Intel Xeon vs. Amazon EC2 Instancesopenssl: RSA 4096-bit Performanceprimesieve: 1e12 Prime Number Generationc-ray: Total Timebuild-llvm: Time To Compilebuild-linux-kernel: Time To Compilex264: H.264 Video Encodingjohn-the-ripper: Blowfishrodinia: OpenMP Streamclusterrodinia: OpenMP LavaMDparboil: OpenMP Stencilparboil: OpenMP LBMnpb: LU.Cnpb: LU.Anpb: EP.CAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138m4.10xlargem4.16xlargec4.4xlargec4.8xlarge3294.5314.082.84175.1137.40292.632955323.1530.9613.8250.9546983.3463042.801610.173306.3714.062.85192.2639.28288.283433514.6630.147.7538.3750046.9062479.081607.684826.7011.812.84134.6430.54310.793037322.5231.447.7550.0450072.2553974.491815.892229.1022.186.45213.9441.84264.842352023.6656.4711.4774.4336509.8637949.85629.383835.7014.394.21147.2430.62304.363903920.2134.197.6347.4543175.9066710.161036.501070.0745.9613.22435.3673.54257.831138225.67115.9112.86114.3118060.6320546.79302.172382.3020.756.08209.7639.89281.332515223.4552.7711.8665.38672.88OpenBenchmarking.org

OpenSSL

RSA 4096-bit Performance

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.0.1gRSA 4096-bit Performance2 x Intel Xeon Gold 6138m4.16xlargeAMD EPYC 7601 (NUMA Interleave All)AMD EPYC 7601c4.8xlargem4.10xlargec4.4xlarge10002000300040005000SE +/- 23.22, N = 3SE +/- 3.53, N = 3SE +/- 13.22, N = 3SE +/- 14.45, N = 3SE +/- 4.10, N = 3SE +/- 1.32, N = 3SE +/- 0.20, N = 34826.703835.703306.373294.532382.302229.101070.071. (CC) gcc options: -m64 -O3 -lssl -lcrypto -ldl

Primesieve

1e12 Prime Number Generation

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 5.4.21e12 Prime Number Generation2 x Intel Xeon Gold 6138AMD EPYC 7601 (NUMA Interleave All)AMD EPYC 7601m4.16xlargec4.8xlargem4.10xlargec4.4xlarge1020304050SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 311.8114.0614.0814.3920.7522.1845.961. (CXX) g++ options: -O2 -fopenmp

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeAMD EPYC 76012 x Intel Xeon Gold 6138AMD EPYC 7601 (NUMA Interleave All)m4.16xlargec4.8xlargem4.10xlargec4.4xlarge3691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 32.842.842.854.216.086.4513.221. (CC) gcc options: -lm -lpthread -O3

Timed LLVM Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 4.0.1Time To Compile2 x Intel Xeon Gold 6138m4.16xlargeAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)c4.8xlargem4.10xlargec4.4xlarge90180270360450SE +/- 0.81, N = 3SE +/- 1.14, N = 3SE +/- 3.02, N = 4SE +/- 2.34, N = 3SE +/- 1.88, N = 3SE +/- 1.94, N = 3SE +/- 1.06, N = 3134.64147.24175.11192.26209.76213.94435.36

Timed Linux Kernel Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 4.9Time To Compile2 x Intel Xeon Gold 6138m4.16xlargeAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)c4.8xlargem4.10xlargec4.4xlarge1632486480SE +/- 0.89, N = 6SE +/- 0.52, N = 6SE +/- 0.51, N = 6SE +/- 0.58, N = 5SE +/- 0.67, N = 4SE +/- 0.63, N = 5SE +/- 0.82, N = 330.5430.6237.4039.2839.8941.8473.54

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2017-09-08H.264 Video Encoding2 x Intel Xeon Gold 6138m4.16xlargeAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)c4.8xlargem4.10xlargec4.4xlarge70140210280350SE +/- 3.63, N = 3SE +/- 3.42, N = 3SE +/- 1.27, N = 3SE +/- 0.45, N = 3SE +/- 5.81, N = 6SE +/- 5.12, N = 3SE +/- 0.88, N = 3310.79304.36292.63288.28281.33264.84257.83-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

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.8.0Test: Blowfishm4.16xlargeAMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138AMD EPYC 7601c4.8xlargem4.10xlargec4.4xlarge8K16K24K32K40KSE +/- 25.17, N = 3SE +/- 764.17, N = 6SE +/- 2076.33, N = 6SE +/- 2253.07, N = 6SE +/- 181.43, N = 3SE +/- 73.10, N = 3390393433530373295532515223520113821. (CC) gcc options: -fopenmp -lcrypt

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP StreamclusterAMD EPYC 7601 (NUMA Interleave All)m4.16xlarge2 x Intel Xeon Gold 6138AMD EPYC 7601c4.8xlargem4.10xlargec4.4xlarge612182430SE +/- 0.27, N = 6SE +/- 1.20, N = 6SE +/- 0.42, N = 3SE +/- 1.43, N = 6SE +/- 1.06, N = 6SE +/- 0.36, N = 3SE +/- 0.53, N = 614.6620.2122.5223.1523.4523.6625.671. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LavaMDAMD EPYC 7601 (NUMA Interleave All)AMD EPYC 76012 x Intel Xeon Gold 6138m4.16xlargec4.8xlargem4.10xlargec4.4xlarge306090120150SE +/- 0.15, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 3SE +/- 0.15, N = 3SE +/- 0.15, N = 3SE +/- 0.17, N = 3SE +/- 0.12, N = 330.1430.9631.4434.1952.7756.47115.911. (CXX) g++ options: -O2 -lOpenCL

Parboil

Test: OpenMP Stencil

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP Stencilm4.16xlargeAMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138m4.10xlargec4.8xlargec4.4xlargeAMD EPYC 760148121620SE +/- 0.15, N = 3SE +/- 0.00, N = 3SE +/- 0.17, N = 6SE +/- 0.08, N = 3SE +/- 0.31, N = 6SE +/- 0.11, N = 3SE +/- 0.68, N = 67.637.757.7511.4711.8612.8613.821. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenMP LBM

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP LBMAMD EPYC 7601 (NUMA Interleave All)m4.16xlarge2 x Intel Xeon Gold 6138AMD EPYC 7601c4.8xlargem4.10xlargec4.4xlarge306090120150SE +/- 0.26, N = 3SE +/- 3.52, N = 6SE +/- 1.14, N = 6SE +/- 0.66, N = 3SE +/- 1.81, N = 6SE +/- 3.88, N = 6SE +/- 0.37, N = 338.3747.4550.0450.9565.3874.43114.311. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

NAS Parallel Benchmarks

Test / Class: LU.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: LU.C2 x Intel Xeon Gold 6138AMD EPYC 7601 (NUMA Interleave All)AMD EPYC 7601m4.16xlargem4.10xlargec4.4xlarge11K22K33K44K55KSE +/- 637.78, N = 3SE +/- 25.06, N = 3SE +/- 747.98, N = 3SE +/- 341.99, N = 3SE +/- 143.51, N = 3SE +/- 20.94, N = 350072.2550046.9046983.3443175.9036509.8618060.631. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 2 x Intel Xeon Gold 6138: Open MPI 2.0.23. AMD EPYC 7601 (NUMA Interleave All): Open MPI 2.0.24. AMD EPYC 7601: Open MPI 2.0.25. m4.16xlarge: Open MPI 1.10.26. m4.10xlarge: Open MPI 1.10.27. c4.4xlarge: Open MPI 1.10.2

NAS Parallel Benchmarks

Test / Class: LU.A

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: LU.Am4.16xlargeAMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)2 x Intel Xeon Gold 6138m4.10xlargec4.4xlarge14K28K42K56K70KSE +/- 637.44, N = 3SE +/- 1051.01, N = 4SE +/- 437.89, N = 3SE +/- 3783.72, N = 6SE +/- 902.21, N = 6SE +/- 22.31, N = 366710.1663042.8062479.0853974.4937949.8520546.791. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. m4.16xlarge: Open MPI 1.10.23. AMD EPYC 7601: Open MPI 2.0.24. AMD EPYC 7601 (NUMA Interleave All): Open MPI 2.0.25. 2 x Intel Xeon Gold 6138: Open MPI 2.0.26. m4.10xlarge: Open MPI 1.10.27. c4.4xlarge: Open MPI 1.10.2

NAS Parallel Benchmarks

Test / Class: EP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: EP.C2 x Intel Xeon Gold 6138AMD EPYC 7601AMD EPYC 7601 (NUMA Interleave All)m4.16xlargec4.8xlargem4.10xlargec4.4xlarge400800120016002000SE +/- 34.76, N = 6SE +/- 0.32, N = 3SE +/- 0.90, N = 3SE +/- 4.41, N = 3SE +/- 1.49, N = 3SE +/- 1.36, N = 3SE +/- 0.85, N = 31815.891610.171607.681036.50672.88629.38302.171. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 2 x Intel Xeon Gold 6138: Open MPI 2.0.23. AMD EPYC 7601: Open MPI 2.0.24. AMD EPYC 7601 (NUMA Interleave All): Open MPI 2.0.25. m4.16xlarge: Open MPI 1.10.26. c4.8xlarge: Open MPI 1.10.27. m4.10xlarge: Open MPI 1.10.28. c4.4xlarge: Open MPI 1.10.2

Phoronix Test Suite v10.8.4