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/1709181-TY-AMAZON80135&sor&gru&export=txt .
AMD EPYC vs. Intel Xeon vs. Amazon EC2 Instances Processor Motherboard Chipset Memory Disk Graphics Monitor Network OS Kernel Display Driver Compiler File-System Screen Resolution Desktop System Layer AMD EPYC 7601 AMD EPYC 7601 (NUMA Interleave All) 2 x Intel Xeon Gold 6138 m4.10xlarge m4.16xlarge c4.4xlarge AMD EPYC 7601 32-Core @ 2.20GHz (64 Cores) TYAN B8026T70AE24HR AMD Device 1450 129024MB 234GB ASPEED ASPEED Family Acer P243W Broadcom Limited NetXtreme BCM5720 Gigabit PCIe Ubuntu 17.04 4.13.0-041300-generic (x86_64) modesetting 1.19.3 GCC 6.3.0 20170406 ext4 1920x1200 Unity 7.5.0 2 x Intel Xeon Gold 6138 @ 3.70GHz (80 Cores) TYAN S7106 Intel Device 2020 96256MB 256GB Samsung SSD 850 + 2000GB Seagate ST2000DM006-2DM1 + 2 x 120GB TOSHIBA-TR150 Intel I210 Gigabit Connection 2 x Intel Xeon E5-2676 v3 @ 3.00GHz (40 Cores) Xen HVM domU Intel 440FX- 82441FX PMC 161792MB 8GB Cirrus Logic GD 5446 Intel 82599 Virtual Function Ubuntu 16.04 4.4.0-1022-aws (x86_64) GCC 5.4.0 20160609 Xen HVM domU 4.2.amazon 2 x Intel Xeon E5-2686 v4 @ 3.00GHz (64 Cores) 258048MB Device 1d0f:ec20 Intel Xeon E5-2666 v3 @ 2.90GHz (16 Cores) 30720MB Intel 82599 Virtual Function OpenBenchmarking.org Compiler 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 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
AMD EPYC vs. Intel Xeon vs. Amazon EC2 Instances x264: H.264 Video Encoding john-the-ripper: Blowfish openssl: RSA 4096-bit Performance npb: EP.C npb: LU.A npb: LU.C parboil: OpenMP LBM parboil: OpenMP Stencil rodinia: OpenMP LavaMD rodinia: OpenMP Streamcluster build-linux-kernel: Time To Compile build-llvm: Time To Compile c-ray: Total Time primesieve: 1e12 Prime Number Generation AMD EPYC 7601 AMD EPYC 7601 (NUMA Interleave All) 2 x Intel Xeon Gold 6138 m4.10xlarge m4.16xlarge c4.4xlarge 292.63 29553 3294.53 1610.17 63042.80 46983.34 50.95 13.82 30.96 23.15 37.40 175.11 2.84 14.08 288.28 34335 3306.37 1607.68 62479.08 50046.90 38.37 7.75 30.14 14.66 39.28 192.26 2.85 14.06 310.79 30373 4826.70 1815.89 53974.49 50072.25 50.04 7.75 31.44 22.52 30.54 134.64 2.84 11.81 264.84 23520 2229.10 629.38 37949.85 36509.86 74.43 11.47 56.47 23.66 41.84 213.94 6.45 22.18 304.36 39039 3835.70 1036.50 66710.16 43175.90 47.45 7.63 34.19 20.21 30.62 147.24 4.21 14.39 257.83 11382 1070.07 302.17 20546.79 18060.63 114.31 12.86 115.91 25.67 73.54 435.36 13.22 45.96 OpenBenchmarking.org
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2017-09-08 H.264 Video Encoding 2 x Intel Xeon Gold 6138 m4.16xlarge AMD EPYC 7601 AMD EPYC 7601 (NUMA Interleave All) m4.10xlarge c4.4xlarge 70 140 210 280 350 SE +/- 3.63, N = 3 SE +/- 3.42, N = 3 SE +/- 1.27, N = 3 SE +/- 0.45, N = 3 SE +/- 5.12, N = 3 SE +/- 0.88, N = 3 310.79 304.36 292.63 288.28 264.84 257.83 -lavformat -lavcodec -lavutil -lswscale -lavformat -lavcodec -lavutil -lswscale 1. (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.org Real C/S, More Is Better John The Ripper 1.8.0 Test: Blowfish m4.16xlarge AMD EPYC 7601 (NUMA Interleave All) 2 x Intel Xeon Gold 6138 AMD EPYC 7601 m4.10xlarge c4.4xlarge 8K 16K 24K 32K 40K SE +/- 25.17, N = 3 SE +/- 764.17, N = 6 SE +/- 2076.33, N = 6 SE +/- 2253.07, N = 6 SE +/- 73.10, N = 3 39039 34335 30373 29553 23520 11382 1. (CC) gcc options: -fopenmp -lcrypt
OpenSSL RSA 4096-bit Performance OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.0.1g RSA 4096-bit Performance 2 x Intel Xeon Gold 6138 m4.16xlarge AMD EPYC 7601 (NUMA Interleave All) AMD EPYC 7601 m4.10xlarge c4.4xlarge 1000 2000 3000 4000 5000 SE +/- 23.22, N = 3 SE +/- 3.53, N = 3 SE +/- 13.22, N = 3 SE +/- 14.45, N = 3 SE +/- 1.32, N = 3 SE +/- 0.20, N = 3 4826.70 3835.70 3306.37 3294.53 2229.10 1070.07 1. (CC) gcc options: -m64 -O3 -lssl -lcrypto -ldl
NAS Parallel Benchmarks Test / Class: EP.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3 Test / Class: EP.C 2 x Intel Xeon Gold 6138 AMD EPYC 7601 AMD EPYC 7601 (NUMA Interleave All) m4.16xlarge m4.10xlarge c4.4xlarge 400 800 1200 1600 2000 SE +/- 34.76, N = 6 SE +/- 0.32, N = 3 SE +/- 0.90, N = 3 SE +/- 4.41, N = 3 SE +/- 1.36, N = 3 SE +/- 0.85, N = 3 1815.89 1610.17 1607.68 1036.50 629.38 302.17 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. 2 x Intel Xeon Gold 6138: Open MPI 2.0.2 3. AMD EPYC 7601: Open MPI 2.0.2 4. AMD EPYC 7601 (NUMA Interleave All): Open MPI 2.0.2 5. m4.16xlarge: Open MPI 1.10.2 6. m4.10xlarge: Open MPI 1.10.2 7. c4.4xlarge: Open MPI 1.10.2
NAS Parallel Benchmarks Test / Class: LU.A OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3 Test / Class: LU.A m4.16xlarge AMD EPYC 7601 AMD EPYC 7601 (NUMA Interleave All) 2 x Intel Xeon Gold 6138 m4.10xlarge c4.4xlarge 14K 28K 42K 56K 70K SE +/- 637.44, N = 3 SE +/- 1051.01, N = 4 SE +/- 437.89, N = 3 SE +/- 3783.72, N = 6 SE +/- 902.21, N = 6 SE +/- 22.31, N = 3 66710.16 63042.80 62479.08 53974.49 37949.85 20546.79 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. m4.16xlarge: Open MPI 1.10.2 3. AMD EPYC 7601: Open MPI 2.0.2 4. AMD EPYC 7601 (NUMA Interleave All): Open MPI 2.0.2 5. 2 x Intel Xeon Gold 6138: Open MPI 2.0.2 6. m4.10xlarge: Open MPI 1.10.2 7. c4.4xlarge: Open MPI 1.10.2
NAS Parallel Benchmarks Test / Class: LU.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3 Test / Class: LU.C 2 x Intel Xeon Gold 6138 AMD EPYC 7601 (NUMA Interleave All) AMD EPYC 7601 m4.16xlarge m4.10xlarge c4.4xlarge 11K 22K 33K 44K 55K SE +/- 637.78, N = 3 SE +/- 25.06, N = 3 SE +/- 747.98, N = 3 SE +/- 341.99, N = 3 SE +/- 143.51, N = 3 SE +/- 20.94, N = 3 50072.25 50046.90 46983.34 43175.90 36509.86 18060.63 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. 2 x Intel Xeon Gold 6138: Open MPI 2.0.2 3. AMD EPYC 7601 (NUMA Interleave All): Open MPI 2.0.2 4. AMD EPYC 7601: Open MPI 2.0.2 5. m4.16xlarge: Open MPI 1.10.2 6. m4.10xlarge: Open MPI 1.10.2 7. c4.4xlarge: Open MPI 1.10.2
Parboil Test: OpenMP LBM OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM AMD EPYC 7601 (NUMA Interleave All) m4.16xlarge 2 x Intel Xeon Gold 6138 AMD EPYC 7601 m4.10xlarge c4.4xlarge 30 60 90 120 150 SE +/- 0.26, N = 3 SE +/- 3.52, N = 6 SE +/- 1.14, N = 6 SE +/- 0.66, N = 3 SE +/- 3.88, N = 6 SE +/- 0.37, N = 3 38.37 47.45 50.04 50.95 74.43 114.31 1. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp
Parboil Test: OpenMP Stencil OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil m4.16xlarge AMD EPYC 7601 (NUMA Interleave All) 2 x Intel Xeon Gold 6138 m4.10xlarge c4.4xlarge AMD EPYC 7601 4 8 12 16 20 SE +/- 0.15, N = 3 SE +/- 0.00, N = 3 SE +/- 0.17, N = 6 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 SE +/- 0.68, N = 6 7.63 7.75 7.75 11.47 12.86 13.82 1. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp
Rodinia Test: OpenMP LavaMD OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD AMD EPYC 7601 (NUMA Interleave All) AMD EPYC 7601 2 x Intel Xeon Gold 6138 m4.16xlarge m4.10xlarge c4.4xlarge 30 60 90 120 150 SE +/- 0.15, N = 3 SE +/- 0.01, N = 3 SE +/- 0.10, N = 3 SE +/- 0.15, N = 3 SE +/- 0.17, N = 3 SE +/- 0.12, N = 3 30.14 30.96 31.44 34.19 56.47 115.91 1. (CXX) g++ options: -O2 -lOpenCL
Rodinia Test: OpenMP Streamcluster OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster AMD EPYC 7601 (NUMA Interleave All) m4.16xlarge 2 x Intel Xeon Gold 6138 AMD EPYC 7601 m4.10xlarge c4.4xlarge 6 12 18 24 30 SE +/- 0.27, N = 6 SE +/- 1.20, N = 6 SE +/- 0.42, N = 3 SE +/- 1.43, N = 6 SE +/- 0.36, N = 3 SE +/- 0.53, N = 6 14.66 20.21 22.52 23.15 23.66 25.67 1. (CXX) g++ options: -O2 -lOpenCL
Timed Linux Kernel Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 4.9 Time To Compile 2 x Intel Xeon Gold 6138 m4.16xlarge AMD EPYC 7601 AMD EPYC 7601 (NUMA Interleave All) m4.10xlarge c4.4xlarge 16 32 48 64 80 SE +/- 0.89, N = 6 SE +/- 0.52, N = 6 SE +/- 0.51, N = 6 SE +/- 0.58, N = 5 SE +/- 0.63, N = 5 SE +/- 0.82, N = 3 30.54 30.62 37.40 39.28 41.84 73.54
Timed LLVM Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 4.0.1 Time To Compile 2 x Intel Xeon Gold 6138 m4.16xlarge AMD EPYC 7601 AMD EPYC 7601 (NUMA Interleave All) m4.10xlarge c4.4xlarge 90 180 270 360 450 SE +/- 0.81, N = 3 SE +/- 1.14, N = 3 SE +/- 3.02, N = 4 SE +/- 2.34, N = 3 SE +/- 1.94, N = 3 SE +/- 1.06, N = 3 134.64 147.24 175.11 192.26 213.94 435.36
C-Ray Total Time OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time AMD EPYC 7601 2 x Intel Xeon Gold 6138 AMD EPYC 7601 (NUMA Interleave All) m4.16xlarge m4.10xlarge c4.4xlarge 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 2.84 2.84 2.85 4.21 6.45 13.22 1. (CC) gcc options: -lm -lpthread -O3
Primesieve 1e12 Prime Number Generation OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 5.4.2 1e12 Prime Number Generation 2 x Intel Xeon Gold 6138 AMD EPYC 7601 (NUMA Interleave All) AMD EPYC 7601 m4.16xlarge m4.10xlarge c4.4xlarge 10 20 30 40 50 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 11.81 14.06 14.08 14.39 22.18 45.96 1. (CXX) g++ options: -O2 -fopenmp
Phoronix Test Suite v10.8.5