Graviton2 versus AMD EPYC 7742 bare metal benchmarks by Michael Larabel.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2005220-NI-GRAVITON200 Amazon Graviton2 vs. AMD EPYC 7742 - Phoronix Test Suite Amazon Graviton2 vs. AMD EPYC 7742 Graviton2 versus AMD EPYC 7742 bare metal benchmarks by Michael Larabel.
HTML result view exported from: https://openbenchmarking.org/result/2005220-NI-GRAVITON200&sor&grs .
Amazon Graviton2 vs. AMD EPYC 7742 Processor Motherboard Memory Disk Network Chipset Graphics Monitor OS Kernel Compiler File-System Desktop Display Server Display Driver OpenGL Screen Resolution Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t ARMv8 Neoverse-N1 (64 Cores) Amazon EC2 m6g.metal v1.0 252GB 137GB Amazon Elastic Block Store Amazon Elastic Ubuntu 20.04 5.4.0-1009-aws (aarch64) GCC 9.3.0 ext4 AMD EPYC 7742 64-Core @ 2.25GHz (64 Cores) AMD DAYTONA_X (RDY1006G BIOS) AMD Starship/Matisse 3841GB Micron_9300_MTFDHAL3T8TDP llvmpipe 252GB VE228 2 x Mellanox MT27710 5.4.0-31-generic (x86_64) GNOME Shell 3.36.1 X Server 1.20.8 modesetting 1.20.8 3.3 Mesa 20.0.4 (LLVM 9.0.1 128 bits) 1920x1080 AMD EPYC 7742 64-Core @ 2.25GHz (64 Cores / 128 Threads) OpenBenchmarking.org Compiler Details - Graviton2 64c: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - EPYC 7742 64c: --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --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=auto --with-tune=generic --without-cuda-driver -v - EPYC 7742 64c 128t: --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --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=auto --with-tune=generic --without-cuda-driver -v Disk Details - Graviton2 64c: NONE / discard,relatime,rw - EPYC 7742 64c: NONE / errors=remount-ro,relatime,rw - EPYC 7742 64c 128t: NONE / errors=remount-ro,relatime,rw Java Details - OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-3ubuntu1) Python Details - Python 2.7.18rc1 + Python 3.8.2 Security Details - Graviton2 64c: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + tsx_async_abort: Not affected - EPYC 7742 64c: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: disabled RSB filling + tsx_async_abort: Not affected - EPYC 7742 64c 128t: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + tsx_async_abort: Not affected Processor Details - EPYC 7742 64c, EPYC 7742 64c 128t: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301034
Amazon Graviton2 vs. AMD EPYC 7742 rav1e: 10 npb: LU.C cloverleaf: Lagrangian-Eulerian Hydrodynamics openssl: RSA 4096-bit Performance mysqlslap: 4 mysqlslap: 8 npb: BT.C mysqlslap: 1 mysqlslap: 16 john-the-ripper: MD5 encode-flac: WAV To FLAC cryptopp: Integer + Elliptic Curve Public Key Algorithms build-linux-kernel: Time To Compile parboil: OpenMP MRI Gridding stress-ng: CPU Stress stress-ng: System V Message Passing basis: UASTC Level 0 mrbayes: Primate Phylogeny Analysis npb: SP.B nettle: poly1305-aes cryptopp: All Algorithms vpxenc: Speed 5 lczero: Eigen mysqlslap: 32 npb: EP.D stress-ng: Glibc C String Functions swet: Average npb: EP.C stress-ng: Atomic cryptopp: Keyed Algorithms stress-ng: Malloc john-the-ripper: Blowfish stress-ng: Semaphores stress-ng: MMAP nettle: aes256 nettle: chacha build-imagemagick: Time To Compile stress-ng: Glibc Qsort Data Sorting minion: Solitaire npb: MG.C coremark: CoreMark Size 666 - Iterations Per Second cython-bench: tjbench: Decompression Throughput rodinia: OpenMP Streamcluster optcarrot: Optimized Benchmark n-queens: Elapsed Time gimp: rotate compress-7zip: Compress Speed Test basis: ETC1S renaissance: Scala Dotty build-php: Time To Compile gimp: auto-levels m-queens: Time To Solve darktable: Server Room - CPU-only gimp: unsharp-mask npb: FT.C gegl: Rotate 90 Degrees blosc: blosclz mysqlslap: 512 gegl: Cartoon minion: Graceful asmfish: 1024 Hash Memory, 26 Depth dav1d: Summer Nature 4K gimp: resize gegl: Scale build2: Time To Compile mysqlslap: 256 redis: LPUSH primesieve: 1e12 Prime Number Generation smallpt: Global Illumination Renderer; 128 Samples himeno: Poisson Pressure Solver stress-ng: Vector Math build-llvm: Time To Compile mysqlslap: 128 renaissance: Apache Spark Bayes node-octane: dav1d: Chimera 1080p c-ray: Total Time - 4K, 16 Rays Per Pixel nettle: sha512 dav1d: Summer Nature 1080p rodinia: OpenMP LavaMD go-benchmark: garbage gegl: Color Enhance cassandra: Writes graphics-magick: Enhanced minion: Quasigroup darktable: Masskrug - CPU-only gegl: Crop parboil: OpenMP LBM mysqlslap: 64 gegl: Reflect hpcg: graphics-magick: HWB Color Space basis: UASTC Level 2 + RDO Post-Processing renaissance: Apache Spark PageRank gromacs: Water Benchmark encode-mp3: WAV To MP3 build-gcc: Time To Compile basis: UASTC Level 3 perl-benchmark: Pod2html graphics-magick: Sharpen node-express-loadtest: tscp: AI Chess Performance basis: UASTC Level 2 sockperf: Throughput darktable: Boat - CPU-only graphics-magick: Swirl renaissance: Rand Forest npb: IS.D cryptopp: Unkeyed Algorithms graphics-magick: Noise-Gaussian stress-ng: Crypto graphics-magick: Rotate gegl: Antialias rodinia: OpenMP CFD Solver x264: H.264 Video Encoding go-benchmark: json rust-mandel: Time To Complete Serial/Parallel Mandelbrot gegl: Tile Glass go-benchmark: build rawtherapee: Total Benchmark Time gegl: Wavelet Blur neat: renaissance: Apache Spark ALS compress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 stress-ng: Context Switching stress-ng: Memory Copying stress-ng: CPU Cache redis: SET redis: GET redis: SADD redis: LPOP pgbench: Buffer Test - Normal Load - Read Write pgbench: Buffer Test - Normal Load - Read Only ebizzy: ttsiod-renderer: Phong Rendering With Soft-Shadow Mapping renaissance: Genetic Algorithm Using Jenetics + Futures renaissance: Akka Unbalanced Cobwebbed Tree renaissance: Savina Reactors.IO java-gradle-perf: Reactor mafft: Multiple Sequence Alignment blogbench: Write Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 0.311 18641.52 1.61 2628.3 1797 1635 24463.10 3086 1531 1275000 32.664 1270.011057 87.912 40.121080 7220.62 10623060.95 19.798 257.320 9935.40 949.75 636.516544 9.02 2318 1288 2228.04 3519134.35 321026102 2236.16 231714.32 310.970923 712364028.71 43335 6276772.99 865.74 2782.98 449.931 28.667 441.76 127.633151 25892.62 1237272.875267 81.590 108.800089 13.999 68.67 3.67 22.087 179794 80.213 9767.832 64.152 26.153 18.743 2.585 31.940 21669.91 67.124 9658.2 550 158.724 71.134013 105936115 239.59 13.246 9.597 85.925 548 963398.88 6.328 2.912 2890.311784 353814.74 300.792 651 2676.428 29340 588.39 15.545 379.99 648.05 8.131 772447 89.185 266989 701 165.655664 3.566 12.106 20.609622 965 45.266 21.4310 1018 878.012 25931.722 2.747 10.903 1014.193 21.251 0.17490547 717 6553 869010 14.958 386634 3.755 1513 5806.623 844.58 344.032817 561 11364.24 496 54.334 8.174 140.78 1395314 43.031 43.055 31136250384 61.395 85.547 18.638 6639.738 9.407 21.909 21475660.36 10027.87 13.37 1244328.34 1644303.92 1340494.50 1755471.58 2280.175134 176350.406937 345427 411.617 2374.701 26628.122 33873.794 491.393 2.246 18017 2.669 109034.36 0.29 12843.0 7943 7055 102257.04 11194 5474 4411667 9.871 4206.418142 29.645 93.320796 16311.72 22845900.42 7.846 105.653 18585.84 2103.68 1388.440315 18.16 4102 2580 4318.05 5116590.96 607550474 4176.38 424049.95 547.790418 1097638087.34 75564 5325535.94 538.85 4794.98 771.233 17.725 565.80 77.439111 42388.33 1536926.324306 50.960 172.621813 9.151 106.94 4.007 14.923 215715 53.544 6576.560 44.093 17.756 19.039 1.749 22.007 31351.48 47.747 9296.5 445 114.121 51.159415 94537908 271.32 9.692 7.080 62.960 444 1150357.75 4.664 2.777 3827.226775 316216.20 224.887 534 3426.461 38597 456.90 12.425 491.47 525.13 6.676 981570 70.242 211622 864 132.636762 2.864 9.789 17.899186 1186 36.898 17.7464 1241 726.973 21687.606 2.501 9.111 856.970 19.287 0.14785560 609 7683 1029878 13.207 454975 3.195 1582 5005.110 967.61 303.239346 582 11923.07 556 48.624 7.676 154.57 1411199 40.428 39.078 28238211426 55.797 77.870 17.075 6202.161 9.003 21.763 15203224.41 14243.42 62.25 1405513.41 1721920.61 1559703.99 1720173.10 42359.001766 611893.068111 2488944 978.007 3206.972 18076.818 16569.162 341.376 2.160 39604 2.324 102048.49 0.40 12586.6 7447 6593 92544.26 11293 5155 4525000 9.798 4195.347303 26.766 110.139532 19526.19 26968236.93 7.855 106.972 22447.14 2102.49 1388.273600 18.18 2041 2481 4271.16 6797662.87 605881667 4129.25 403796.58 547.492068 1248712298.23 74914 9258346.82 934.59 4795.26 771.141 17.074 741.19 77.673532 38200.26 1988375.158409 51.379 172.617150 8.936 107.11 2.602 14.705 269626 53.661 6532.497 42.946 17.638 12.871 1.995 22.218 31228.98 47.926 6876.1 394 114.389 51.443073 131314356 331.42 9.661 7.002 64.256 403 1307306.84 4.721 2.150 3912.005207 424666.61 226.620 487 3529.771 38332 599.05 11.920 491.16 677.59 6.319 895616 70.360 220011 884 132.610696 3.199 9.782 22.022019 1133 37.037 17.5552 1174 728.889 21483.402 2.289 9.115 850.033 17.828 0.14684740 602 7771 1022643 12.666 448694 3.492 1764 5011.822 971.93 303.112596 634 12828.13 550 48.748 7.350 156.16 1273253 39.004 39.045 29667943541 58.835 78.115 17.277 6324.940 9.067 21.896 26364845.16 12405.95 46.80 1493004.17 2047459.14 1683866.41 2168678.10 37417.906497 988939.198432 3075320 881.381 3143.389 21717.100 19664.723 340.723 2.136 48646 OpenBenchmarking.org
rav1e Speed: 10 OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.3.0 Speed: 10 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 0.6005 1.201 1.8015 2.402 3.0025 SE +/- 0.005, N = 3 SE +/- 0.005, N = 3 SE +/- 0.000, N = 3 2.669 2.324 0.311
NAS Parallel Benchmarks Test / Class: LU.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 20K 40K 60K 80K 100K SE +/- 113.11, N = 3 SE +/- 1012.10, N = 15 SE +/- 140.01, N = 3 109034.36 102048.49 18641.52 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
CloverLeaf Lagrangian-Eulerian Hydrodynamics OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 0.3623 0.7246 1.0869 1.4492 1.8115 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 0.29 0.40 1.61 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
OpenSSL RSA 4096-bit Performance OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 3K 6K 9K 12K 15K SE +/- 12.96, N = 3 SE +/- 0.69, N = 3 SE +/- 0.56, N = 3 12843.0 12586.6 2628.3 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
MariaDB Clients: 4 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 4 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 2K 4K 6K 8K 10K SE +/- 6.35, N = 10 SE +/- 75.43, N = 8 SE +/- 17.47, N = 3 7943 7447 1797 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
MariaDB Clients: 8 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 8 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 1500 3000 4500 6000 7500 SE +/- 12.49, N = 4 SE +/- 14.52, N = 3 SE +/- 6.23, N = 3 7055 6593 1635 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
NAS Parallel Benchmarks Test / Class: BT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 20K 40K 60K 80K 100K SE +/- 1483.41, N = 4 SE +/- 1221.04, N = 5 SE +/- 4.67, N = 3 102257.04 92544.26 24463.10 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
MariaDB Clients: 1 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 1 EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 2K 4K 6K 8K 10K SE +/- 137.28, N = 15 SE +/- 41.61, N = 3 SE +/- 9.56, N = 3 11293 11194 3086 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
MariaDB Clients: 16 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 16 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 1200 2400 3600 4800 6000 SE +/- 10.01, N = 3 SE +/- 26.71, N = 3 SE +/- 9.85, N = 3 5474 5155 1531 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
John The Ripper Test: MD5 OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: MD5 EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 1000K 2000K 3000K 4000K 5000K SE +/- 4163.33, N = 3 SE +/- 2666.67, N = 3 SE +/- 3000.00, N = 3 4525000 4411667 1275000 -m64 -m64 1. (CC) gcc options: -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 8 16 24 32 40 SE +/- 0.008, N = 5 SE +/- 0.006, N = 5 SE +/- 0.025, N = 5 9.798 9.871 32.664 1. (CXX) g++ options: -O2 -fvisibility=hidden -logg -lm
Crypto++ Test: Integer + Elliptic Curve Public Key Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Integer + Elliptic Curve Public Key Algorithms EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 900 1800 2700 3600 4500 SE +/- 2.97, N = 3 SE +/- 2.27, N = 3 SE +/- 0.65, N = 3 4206.42 4195.35 1270.01 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
Timed Linux Kernel Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.4 Time To Compile EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 20 40 60 80 100 SE +/- 0.20, N = 13 SE +/- 0.28, N = 10 SE +/- 0.99, N = 3 26.77 29.65 87.91
Parboil Test: OpenMP MRI Gridding OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 20 40 60 80 100 SE +/- 0.24, N = 3 SE +/- 0.25, N = 3 SE +/- 0.32, N = 3 40.12 93.32 110.14 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Stress-NG Test: CPU Stress OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: CPU Stress EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 4K 8K 12K 16K 20K SE +/- 17.20, N = 3 SE +/- 16.32, N = 3 SE +/- 0.34, N = 3 19526.19 16311.72 7220.62 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Stress-NG Test: System V Message Passing OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: System V Message Passing EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 6M 12M 18M 24M 30M SE +/- 373815.20, N = 15 SE +/- 174925.99, N = 3 SE +/- 98960.78, N = 3 26968236.93 22845900.42 10623060.95 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Basis Universal Settings: UASTC Level 0 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 0 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 5 10 15 20 25 SE +/- 0.011, N = 3 SE +/- 0.006, N = 3 SE +/- 0.218, N = 7 7.846 7.855 19.798 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Timed MrBayes Analysis Primate Phylogeny Analysis OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 60 120 180 240 300 SE +/- 0.12, N = 3 SE +/- 0.84, N = 3 SE +/- 0.26, N = 3 105.65 106.97 257.32 -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm 1. (CC) gcc options: -O3 -std=c99 -pedantic -lm
NAS Parallel Benchmarks Test / Class: SP.B OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 5K 10K 15K 20K 25K SE +/- 217.12, N = 3 SE +/- 233.51, N = 3 SE +/- 4.95, N = 3 22447.14 18585.84 9935.40 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Nettle Test: poly1305-aes OpenBenchmarking.org Mbyte/s, More Is Better Nettle 3.5.1 Test: poly1305-aes EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 500 1000 1500 2000 2500 SE +/- 0.04, N = 3 SE +/- 0.20, N = 3 SE +/- 1.97, N = 3 2103.68 2102.49 949.75 1. (CC) gcc options: -O2 -ggdb3 -lnettle -lgmp -lm -lcrypto
Crypto++ Test: All Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: All Algorithms EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 300 600 900 1200 1500 SE +/- 1.06, N = 3 SE +/- 0.31, N = 3 SE +/- 0.33, N = 3 1388.44 1388.27 636.52 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
VP9 libvpx Encoding Speed: Speed 5 OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.2 Speed: Speed 5 EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 18.18 18.16 9.02 -m64 -m64 -march=armv8-a 1. (CXX) g++ options: -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.25 Backend: Eigen EPYC 7742 64c Graviton2 64c EPYC 7742 64c 128t 900 1800 2700 3600 4500 SE +/- 38.67, N = 3 SE +/- 39.54, N = 3 SE +/- 25.11, N = 5 4102 2318 2041 1. (CXX) g++ options: -pthread
MariaDB Clients: 32 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 32 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 600 1200 1800 2400 3000 SE +/- 4.43, N = 3 SE +/- 6.17, N = 3 SE +/- 7.72, N = 3 2580 2481 1288 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
NAS Parallel Benchmarks Test / Class: EP.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 900 1800 2700 3600 4500 SE +/- 3.01, N = 3 SE +/- 8.82, N = 3 SE +/- 2.85, N = 3 4318.05 4271.16 2228.04 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Stress-NG Test: Glibc C String Functions OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Glibc C String Functions EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 1.5M 3M 4.5M 6M 7.5M SE +/- 108910.76, N = 3 SE +/- 34550.61, N = 3 SE +/- 3159.35, N = 3 6797662.87 5116590.96 3519134.35 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Swet Average OpenBenchmarking.org Operations Per Second, More Is Better Swet 1.5.16 Average EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 130M 260M 390M 520M 650M SE +/- 4322060.32, N = 3 SE +/- 1731105.92, N = 3 SE +/- 310568.82, N = 3 607550474 605881667 321026102 1. (CC) gcc options: -lm -lpthread -lcurses -lrt
NAS Parallel Benchmarks Test / Class: EP.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 900 1800 2700 3600 4500 SE +/- 23.64, N = 3 SE +/- 21.28, N = 3 SE +/- 1.31, N = 3 4176.38 4129.25 2236.16 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Stress-NG Test: Atomic OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Atomic EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 90K 180K 270K 360K 450K SE +/- 146.09, N = 3 SE +/- 147.04, N = 3 SE +/- 756.68, N = 3 424049.95 403796.58 231714.32 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Crypto++ Test: Keyed Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Keyed Algorithms EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 120 240 360 480 600 SE +/- 0.27, N = 3 SE +/- 0.08, N = 3 SE +/- 0.15, N = 3 547.79 547.49 310.97 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
Stress-NG Test: Malloc OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Malloc EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 300M 600M 900M 1200M 1500M SE +/- 2770486.87, N = 3 SE +/- 1348531.50, N = 3 SE +/- 603865.65, N = 3 1248712298.23 1097638087.34 712364028.71 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: Blowfish EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 16K 32K 48K 64K 80K SE +/- 17.07, N = 3 SE +/- 49.17, N = 3 SE +/- 11.29, N = 3 75564 74914 43335 -m64 -m64 1. (CC) gcc options: -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Semaphores EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 2M 4M 6M 8M 10M SE +/- 3849.76, N = 3 SE +/- 9411.48, N = 3 SE +/- 6529.14, N = 3 9258346.82 6276772.99 5325535.94 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Stress-NG Test: MMAP OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: MMAP EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 200 400 600 800 1000 SE +/- 0.88, N = 3 SE +/- 0.63, N = 3 SE +/- 0.56, N = 3 934.59 865.74 538.85 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Nettle Test: aes256 OpenBenchmarking.org Mbyte/s, More Is Better Nettle 3.5.1 Test: aes256 EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 1000 2000 3000 4000 5000 SE +/- 0.74, N = 3 SE +/- 0.65, N = 3 SE +/- 0.65, N = 3 4795.26 4794.98 2782.98 MIN: 3477.09 / MAX: 7440.83 MIN: 3477.92 / MAX: 7440.81 MIN: 1656.91 / MAX: 5033.12 1. (CC) gcc options: -O2 -ggdb3 -lnettle -lgmp -lm -lcrypto
Nettle Test: chacha OpenBenchmarking.org Mbyte/s, More Is Better Nettle 3.5.1 Test: chacha EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 170 340 510 680 850 SE +/- 0.07, N = 3 SE +/- 0.13, N = 3 SE +/- 0.52, N = 3 771.23 771.14 449.93 MIN: 393.13 / MAX: 2103.79 MIN: 393.01 / MAX: 2103.79 MIN: 267.62 / MAX: 976 1. (CC) gcc options: -O2 -ggdb3 -lnettle -lgmp -lm -lcrypto
Timed ImageMagick Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 7 14 21 28 35 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.13, N = 3 17.07 17.73 28.67
Stress-NG Test: Glibc Qsort Data Sorting OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Glibc Qsort Data Sorting EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 160 320 480 640 800 SE +/- 1.48, N = 3 SE +/- 8.12, N = 3 SE +/- 0.98, N = 3 741.19 565.80 441.76 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Minion Benchmark: Solitaire OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Solitaire EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 30 60 90 120 150 SE +/- 0.34, N = 3 SE +/- 0.34, N = 3 SE +/- 0.31, N = 3 77.44 77.67 127.63 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
NAS Parallel Benchmarks Test / Class: MG.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 9K 18K 27K 36K 45K SE +/- 114.33, N = 3 SE +/- 257.15, N = 3 SE +/- 2.41, N = 3 42388.33 38200.26 25892.62 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Coremark CoreMark Size 666 - Iterations Per Second OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 400K 800K 1200K 1600K 2000K SE +/- 5055.38, N = 3 SE +/- 7872.26, N = 3 SE +/- 138.10, N = 3 1988375.16 1536926.32 1237272.88 1. (CC) gcc options: -O2 -lrt" -lrt
Cython benchmark OpenBenchmarking.org Seconds, Fewer Is Better Cython benchmark 0.27 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 20 40 60 80 100 SE +/- 0.14, N = 3 SE +/- 0.65, N = 3 SE +/- 0.16, N = 3 50.96 51.38 81.59
libjpeg-turbo tjbench Test: Decompression Throughput OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 2.0.2 Test: Decompression Throughput EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 40 80 120 160 200 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 172.62 172.62 108.80 1. (CC) gcc options: -O3 -rdynamic
Rodinia Test: OpenMP Streamcluster OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 4 8 12 16 20 SE +/- 0.010, N = 3 SE +/- 0.024, N = 3 SE +/- 0.146, N = 14 8.936 9.151 13.999 1. (CXX) g++ options: -O2 -lOpenCL
Optcarrot Optimized Benchmark OpenBenchmarking.org FPS, More Is Better Optcarrot Optimized Benchmark EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 20 40 60 80 100 SE +/- 0.45, N = 4 SE +/- 0.34, N = 4 SE +/- 0.35, N = 4 107.11 106.94 68.67 1. EPYC 7742 64c 128t: ruby 2.7.0p0 (2019-12-25 revision 647ee6f091) [x86_64-linux-gnu] 2. EPYC 7742 64c: ruby 2.7.0p0 (2019-12-25 revision 647ee6f091) [x86_64-linux-gnu] 3. Graviton2 64c: ruby 2.7.0p0 (2019-12-25 revision 647ee6f091) [aarch64-linux-gnu]
N-Queens Elapsed Time OpenBenchmarking.org Seconds, Fewer Is Better N-Queens 1.0 Elapsed Time EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 0.9016 1.8032 2.7048 3.6064 4.508 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 2.602 3.670 4.007 1. (CC) gcc options: -static -fopenmp -O3 -march=native
GIMP Test: rotate OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: rotate EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.15, N = 3 SE +/- 0.01, N = 3 14.71 14.92 22.09
7-Zip Compression Compress Speed Test OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 60K 120K 180K 240K 300K SE +/- 620.36, N = 3 SE +/- 477.98, N = 3 SE +/- 597.59, N = 3 269626 215715 179794 1. (CXX) g++ options: -pipe -lpthread
Basis Universal Settings: ETC1S OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: ETC1S EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 20 40 60 80 100 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.43, N = 3 53.54 53.66 80.21 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Renaissance Test: Scala Dotty OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Scala Dotty EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 2K 4K 6K 8K 10K SE +/- 21.98, N = 5 SE +/- 16.73, N = 5 SE +/- 54.25, N = 5 6532.50 6576.56 9767.83
Timed PHP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.4.2 Time To Compile EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 14 28 42 56 70 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 42.95 44.09 64.15
GIMP Test: auto-levels OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: auto-levels EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 6 12 18 24 30 SE +/- 0.13, N = 3 SE +/- 0.16, N = 3 SE +/- 0.15, N = 3 17.64 17.76 26.15
m-queens Time To Solve OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 12.87 18.74 19.04 1. (CXX) g++ options: -fopenmp -O2 -march=native
Darktable Test: Server Room - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.0.1 Test: Server Room - Acceleration: CPU-only EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 0.5816 1.1632 1.7448 2.3264 2.908 SE +/- 0.004, N = 3 SE +/- 0.004, N = 3 SE +/- 0.004, N = 3 1.749 1.995 2.585
GIMP Test: unsharp-mask OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: unsharp-mask EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 7 14 21 28 35 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.23, N = 3 22.01 22.22 31.94
NAS Parallel Benchmarks Test / Class: FT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 7K 14K 21K 28K 35K SE +/- 128.22, N = 3 SE +/- 108.47, N = 3 SE +/- 4.05, N = 3 31351.48 31228.98 21669.91 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
GEGL Operation: Rotate 90 Degrees OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Rotate 90 Degrees EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 15 30 45 60 75 SE +/- 0.15, N = 3 SE +/- 0.02, N = 3 SE +/- 0.10, N = 3 47.75 47.93 67.12
C-Blosc Compressor: blosclz OpenBenchmarking.org MB/s, More Is Better C-Blosc 2.0 Beta 5 Compressor: blosclz Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 2K 4K 6K 8K 10K SE +/- 47.95, N = 3 SE +/- 22.77, N = 3 SE +/- 4.44, N = 3 9658.2 9296.5 6876.1 1. (CXX) g++ options: -rdynamic
MariaDB Clients: 512 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 512 Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 120 240 360 480 600 SE +/- 2.06, N = 3 SE +/- 0.61, N = 3 SE +/- 1.72, N = 3 550 445 394 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
GEGL Operation: Cartoon OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Cartoon EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 40 80 120 160 200 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.18, N = 3 114.12 114.39 158.72
Minion Benchmark: Graceful OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Graceful EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 16 32 48 64 80 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 51.16 51.44 71.13 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
asmFish 1024 Hash Memory, 26 Depth OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 30M 60M 90M 120M 150M SE +/- 1280575.13, N = 3 SE +/- 734052.82, N = 3 SE +/- 775749.19, N = 3 131314356 105936115 94537908
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better dav1d 0.6.0 Video Input: Summer Nature 4K EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 70 140 210 280 350 SE +/- 1.58, N = 3 SE +/- 0.32, N = 3 SE +/- 0.72, N = 3 331.42 271.32 239.59 MIN: 118.03 / MAX: 366.24 MIN: 149.06 / MAX: 289.04 MIN: 138.3 / MAX: 258.62 1. (CC) gcc options: -pthread
GIMP Test: resize OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: resize EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 3 6 9 12 15 SE +/- 0.030, N = 3 SE +/- 0.067, N = 3 SE +/- 0.082, N = 3 9.661 9.692 13.246
GEGL Operation: Scale OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Scale EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 3 6 9 12 15 SE +/- 0.068, N = 3 SE +/- 0.036, N = 3 SE +/- 0.149, N = 3 7.002 7.080 9.597
Build2 Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.12 Time To Compile EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.13, N = 3 SE +/- 0.18, N = 3 62.96 64.26 85.93
MariaDB Clients: 256 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 256 Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 120 240 360 480 600 SE +/- 3.38, N = 3 SE +/- 0.24, N = 3 SE +/- 0.64, N = 3 548 444 403 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
Redis Test: LPUSH OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: LPUSH EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 300K 600K 900K 1200K 1500K SE +/- 21281.45, N = 3 SE +/- 5408.33, N = 3 SE +/- 1934.63, N = 3 1307306.84 1150357.75 963398.88 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Primesieve 1e12 Prime Number Generation OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 7.4 1e12 Prime Number Generation EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 2 4 6 8 10 SE +/- 0.003, N = 3 SE +/- 0.008, N = 3 SE +/- 0.003, N = 3 4.664 4.721 6.328 1. (CXX) g++ options: -O3 -lpthread
Smallpt Global Illumination Renderer; 128 Samples OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 128 Samples EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 0.6552 1.3104 1.9656 2.6208 3.276 SE +/- 0.004, N = 3 SE +/- 0.005, N = 3 SE +/- 0.004, N = 3 2.150 2.777 2.912 1. (CXX) g++ options: -fopenmp -O3
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 800 1600 2400 3200 4000 SE +/- 56.90, N = 4 SE +/- 28.36, N = 3 SE +/- 12.32, N = 3 3912.01 3827.23 2890.31 -mavx2 -mavx2 1. (CC) gcc options: -O3
Stress-NG Test: Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Vector Math EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 90K 180K 270K 360K 450K SE +/- 22.32, N = 3 SE +/- 2.26, N = 3 SE +/- 25.04, N = 3 424666.61 353814.74 316216.20 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Timed LLVM Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 10.0 Time To Compile EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 70 140 210 280 350 SE +/- 1.52, N = 3 SE +/- 0.27, N = 3 SE +/- 4.33, N = 3 224.89 226.62 300.79
MariaDB Clients: 128 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 128 Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 140 280 420 560 700 SE +/- 3.08, N = 3 SE +/- 0.66, N = 3 SE +/- 0.29, N = 3 651 534 487 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
Renaissance Test: Apache Spark Bayes OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Apache Spark Bayes Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 800 1600 2400 3200 4000 SE +/- 16.86, N = 5 SE +/- 31.10, N = 5 SE +/- 31.46, N = 25 2676.43 3426.46 3529.77
Node.js Octane Benchmark OpenBenchmarking.org Score, More Is Better Node.js Octane Benchmark EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 8K 16K 24K 32K 40K SE +/- 54.76, N = 3 SE +/- 155.81, N = 3 SE +/- 82.34, N = 3 38597 38332 29340 1. Nodejs
v10.19.0
dav1d Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.6.0 Video Input: Chimera 1080p EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 130 260 390 520 650 SE +/- 0.89, N = 3 SE +/- 0.55, N = 3 SE +/- 1.39, N = 3 599.05 588.39 456.90 MIN: 345.23 / MAX: 743.97 MIN: 427.39 / MAX: 777.07 MIN: 319.39 / MAX: 559.29 1. (CC) gcc options: -pthread
C-Ray Total Time - 4K, 16 Rays Per Pixel OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 11.92 12.43 15.55 1. (CC) gcc options: -lm -lpthread -O3
Nettle Test: sha512 OpenBenchmarking.org Mbyte/s, More Is Better Nettle 3.5.1 Test: sha512 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 110 220 330 440 550 SE +/- 0.18, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 491.47 491.16 379.99 1. (CC) gcc options: -O2 -ggdb3 -lnettle -lgmp -lm -lcrypto
dav1d Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.6.0 Video Input: Summer Nature 1080p EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 150 300 450 600 750 SE +/- 1.47, N = 3 SE +/- 1.42, N = 3 SE +/- 2.15, N = 3 677.59 648.05 525.13 MIN: 218.5 / MAX: 776.16 MIN: 367.46 / MAX: 714.91 MIN: 291.81 / MAX: 581.55 1. (CC) gcc options: -pthread
Rodinia Test: OpenMP LavaMD OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 2 4 6 8 10 SE +/- 0.006, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 6.319 6.676 8.131 1. (CXX) g++ options: -O2 -lOpenCL
Go Benchmarks Test: garbage OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: garbage Graviton2 64c EPYC 7742 64c 128t EPYC 7742 64c 200K 400K 600K 800K 1000K SE +/- 10663.19, N = 4 SE +/- 12148.68, N = 3 SE +/- 11062.32, N = 3 772447 895616 981570
GEGL Operation: Color Enhance OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Color Enhance EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 20 40 60 80 100 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 70.24 70.36 89.19
Apache Cassandra Test: Writes OpenBenchmarking.org Op/s, More Is Better Apache Cassandra 3.11.4 Test: Writes Graviton2 64c EPYC 7742 64c 128t EPYC 7742 64c 60K 120K 180K 240K 300K SE +/- 4840.50, N = 2 SE +/- 2670.22, N = 3 SE +/- 2526.79, N = 3 266989 220011 211622
GraphicsMagick Operation: Enhanced OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Enhanced EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 200 400 600 800 1000 SE +/- 0.58, N = 3 884 864 701 -lfreetype 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Minion Benchmark: Quasigroup OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Quasigroup EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 40 80 120 160 200 SE +/- 0.29, N = 3 SE +/- 0.21, N = 3 SE +/- 0.84, N = 3 132.61 132.64 165.66 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Darktable Test: Masskrug - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.0.1 Test: Masskrug - Acceleration: CPU-only EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 0.8024 1.6048 2.4072 3.2096 4.012 SE +/- 0.001, N = 3 SE +/- 0.005, N = 3 SE +/- 0.015, N = 3 2.864 3.199 3.566
GEGL Operation: Crop OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Crop EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 3 6 9 12 15 SE +/- 0.018, N = 3 SE +/- 0.066, N = 3 SE +/- 0.021, N = 3 9.782 9.789 12.106
Parboil Test: OpenMP LBM OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM EPYC 7742 64c Graviton2 64c EPYC 7742 64c 128t 5 10 15 20 25 SE +/- 0.28, N = 3 SE +/- 0.03, N = 3 SE +/- 0.29, N = 3 17.90 20.61 22.02 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
MariaDB Clients: 64 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.5.2 Clients: 64 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 300 600 900 1200 1500 SE +/- 0.35, N = 4 SE +/- 1.55, N = 3 SE +/- 6.47, N = 3 1186 1133 965 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl
GEGL Operation: Reflect OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Reflect EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 10 20 30 40 50 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 36.90 37.04 45.27
High Performance Conjugate Gradient OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 5 10 15 20 25 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 21.43 17.75 17.56 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: HWB Color Space EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 300 600 900 1200 1500 SE +/- 1.76, N = 3 SE +/- 3.51, N = 3 SE +/- 1.76, N = 3 1241 1174 1018 -lfreetype 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Basis Universal Settings: UASTC Level 2 + RDO Post-Processing OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 2 + RDO Post-Processing EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 200 400 600 800 1000 SE +/- 0.06, N = 3 SE +/- 0.18, N = 3 SE +/- 0.17, N = 3 726.97 728.89 878.01 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Renaissance Test: Apache Spark PageRank OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Apache Spark PageRank EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 6K 12K 18K 24K 30K SE +/- 241.74, N = 5 SE +/- 125.46, N = 5 SE +/- 278.60, N = 5 21483.40 21687.61 25931.72
GROMACS Water Benchmark OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2020.1 Water Benchmark Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 0.6181 1.2362 1.8543 2.4724 3.0905 SE +/- 0.006, N = 3 SE +/- 0.013, N = 3 SE +/- 0.026, N = 3 2.747 2.501 2.289 1. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 3 6 9 12 15 SE +/- 0.001, N = 3 SE +/- 0.007, N = 3 SE +/- 0.002, N = 3 9.111 9.115 10.903 1. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lncurses -lm
Timed GCC Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GCC Compilation 9.3.0 Time To Compile EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 200 400 600 800 1000 SE +/- 0.17, N = 3 SE +/- 1.11, N = 3 SE +/- 2.04, N = 3 850.03 856.97 1014.19
Basis Universal Settings: UASTC Level 3 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 3 EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 17.83 19.29 21.25 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Perl Benchmarks Test: Pod2html OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Pod2html EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 0.0394 0.0788 0.1182 0.1576 0.197 SE +/- 0.00125382, N = 3 SE +/- 0.00100736, N = 3 SE +/- 0.00053055, N = 3 0.14684740 0.14785560 0.17490547
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Sharpen Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 150 300 450 600 750 SE +/- 0.33, N = 3 717 609 602 -lfreetype 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Node.js Express HTTP Load Test OpenBenchmarking.org Requests Per Second, More Is Better Node.js Express HTTP Load Test EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 1700 3400 5100 6800 8500 SE +/- 90.06, N = 3 SE +/- 15.62, N = 3 SE +/- 79.37, N = 15 7771 7683 6553 1. Nodejs
v10.19.0
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 200K 400K 600K 800K 1000K SE +/- 720.27, N = 5 SE +/- 2923.69, N = 5 SE +/- 512.87, N = 5 1029878 1022643 869010 1. (CC) gcc options: -O3 -march=native
Basis Universal Settings: UASTC Level 2 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 2 EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 12.67 13.21 14.96 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Sockperf Test: Throughput OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.4 Test: Throughput EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 100K 200K 300K 400K 500K SE +/- 4138.37, N = 5 SE +/- 4022.67, N = 5 SE +/- 2029.76, N = 5 454975 448694 386634 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Darktable Test: Boat - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.0.1 Test: Boat - Acceleration: CPU-only EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 0.8449 1.6898 2.5347 3.3796 4.2245 SE +/- 0.016, N = 3 SE +/- 0.017, N = 3 SE +/- 0.004, N = 3 3.195 3.492 3.755
GraphicsMagick Operation: Swirl OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Swirl EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 400 800 1200 1600 2000 SE +/- 5.57, N = 3 SE +/- 1.20, N = 3 SE +/- 3.61, N = 3 1764 1582 1513 -lfreetype 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Renaissance Test: Random Forest OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Random Forest EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 1200 2400 3600 4800 6000 SE +/- 29.52, N = 5 SE +/- 67.01, N = 5 SE +/- 38.49, N = 5 5005.11 5011.82 5806.62
NAS Parallel Benchmarks Test / Class: IS.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 200 400 600 800 1000 SE +/- 14.61, N = 3 SE +/- 12.97, N = 3 SE +/- 3.48, N = 3 971.93 967.61 844.58 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Crypto++ Test: Unkeyed Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Unkeyed Algorithms Graviton2 64c EPYC 7742 64c EPYC 7742 64c 128t 70 140 210 280 350 SE +/- 0.12, N = 3 SE +/- 0.05, N = 3 SE +/- 0.14, N = 3 344.03 303.24 303.11 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
GraphicsMagick Operation: Noise-Gaussian OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Noise-Gaussian EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 140 280 420 560 700 SE +/- 2.40, N = 3 SE +/- 0.88, N = 3 SE +/- 0.67, N = 3 634 582 561 -lfreetype 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Stress-NG Test: Crypto OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Crypto EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 3K 6K 9K 12K 15K SE +/- 29.74, N = 3 SE +/- 18.74, N = 3 SE +/- 25.31, N = 3 12828.13 11923.07 11364.24 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Rotate EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 120 240 360 480 600 SE +/- 4.41, N = 3 SE +/- 2.67, N = 3 SE +/- 6.23, N = 3 556 550 496 -lfreetype 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
GEGL Operation: Antialias OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Antialias EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 12 24 36 48 60 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.13, N = 3 48.62 48.75 54.33
Rodinia Test: OpenMP CFD Solver OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 2 4 6 8 10 SE +/- 0.092, N = 4 SE +/- 0.080, N = 3 SE +/- 0.052, N = 3 7.350 7.676 8.174 1. (CXX) g++ options: -O2 -lOpenCL
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2019-12-17 H.264 Video Encoding EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 30 60 90 120 150 SE +/- 0.23, N = 3 SE +/- 2.58, N = 3 SE +/- 0.35, N = 3 156.16 154.57 140.78 -m64 -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize -m64 -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize -lavformat -lavcodec -lavutil -lswscale 1. (CC) gcc options: -ldl -lm -lpthread
Go Benchmarks Test: json OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: json EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 300K 600K 900K 1200K 1500K SE +/- 16777.18, N = 4 SE +/- 5211.56, N = 3 SE +/- 5729.12, N = 3 1273253 1395314 1411199
Rust Mandelbrot Time To Complete Serial/Parallel Mandelbrot OpenBenchmarking.org Seconds, Fewer Is Better Rust Mandelbrot Time To Complete Serial/Parallel Mandelbrot EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 10 20 30 40 50 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 39.00 40.43 43.03 -m64 -m64 1. (CC) gcc options: -pie -nodefaultlibs -lutil -ldl -lrt -lpthread -lgcc_s -lc -lm
GEGL Operation: Tile Glass OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Tile Glass EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 10 20 30 40 50 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 39.05 39.08 43.06
Go Benchmarks Test: build OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: build EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 7000M 14000M 21000M 28000M 35000M SE +/- 179022572.52, N = 3 SE +/- 53807658.06, N = 3 SE +/- 11752689.24, N = 3 28238211426 29667943541 31136250384
RawTherapee Total Benchmark Time OpenBenchmarking.org Seconds, Fewer Is Better RawTherapee Total Benchmark Time EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 14 28 42 56 70 SE +/- 0.05, N = 3 SE +/- 0.12, N = 3 SE +/- 0.09, N = 3 55.80 58.84 61.40 1. RawTherapee, version 5.8, command line.
GEGL Operation: Wavelet Blur OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Wavelet Blur EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.33, N = 3 77.87 78.12 85.55
Nebular Empirical Analysis Tool OpenBenchmarking.org Seconds, Fewer Is Better Nebular Empirical Analysis Tool 2020-02-29 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 5 10 15 20 25 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 17.08 17.28 18.64 1. (F9X) gfortran options: -cpp -ffree-line-length-0 -Jsource/ -fopenmp -O3 -fno-backtrace
Renaissance Test: Apache Spark ALS OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Apache Spark ALS EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 1400 2800 4200 5600 7000 SE +/- 45.99, N = 16 SE +/- 68.32, N = 5 SE +/- 47.97, N = 5 6202.16 6324.94 6639.74
Zstd Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 3 6 9 12 15 SE +/- 0.153, N = 3 SE +/- 0.007, N = 3 SE +/- 0.002, N = 3 9.003 9.067 9.407 1. (CC) gcc options: -O3 -pthread -lz -llzma
XZ Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 5 10 15 20 25 SE +/- 0.05, N = 3 SE +/- 0.15, N = 3 SE +/- 0.01, N = 3 21.76 21.90 21.91 1. (CC) gcc options: -pthread -fvisibility=hidden -O2
Stress-NG Test: Context Switching OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Context Switching EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 6M 12M 18M 24M 30M SE +/- 714410.24, N = 15 SE +/- 104061.74, N = 3 SE +/- 205500.78, N = 3 26364845.16 21475660.36 15203224.41 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Stress-NG Test: Memory Copying OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Memory Copying EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 3K 6K 9K 12K 15K SE +/- 386.12, N = 15 SE +/- 33.48, N = 3 SE +/- 1.90, N = 3 14243.42 12405.95 10027.87 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Stress-NG Test: CPU Cache OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: CPU Cache EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 14 28 42 56 70 SE +/- 0.90, N = 15 SE +/- 1.15, N = 15 SE +/- 0.57, N = 12 62.25 46.80 13.37 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
Redis Test: SET OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: SET EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 300K 600K 900K 1200K 1500K SE +/- 18650.88, N = 3 SE +/- 23286.40, N = 15 SE +/- 4424.59, N = 3 1493004.17 1405513.41 1244328.34 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: GET EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 400K 800K 1200K 1600K 2000K SE +/- 46774.40, N = 12 SE +/- 49844.86, N = 12 SE +/- 19731.78, N = 3 2047459.14 1721920.61 1644303.92 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Redis Test: SADD OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: SADD EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 400K 800K 1200K 1600K 2000K SE +/- 30991.56, N = 15 SE +/- 37698.60, N = 15 SE +/- 20080.65, N = 3 1683866.41 1559703.99 1340494.50 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Redis Test: LPOP OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: LPOP EPYC 7742 64c 128t Graviton2 64c EPYC 7742 64c 500K 1000K 1500K 2000K 2500K SE +/- 45135.23, N = 15 SE +/- 7219.96, N = 3 SE +/- 87395.86, N = 12 2168678.10 1755471.58 1720173.10 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 9K 18K 27K 36K 45K SE +/- 37.91, N = 3 SE +/- 110.45, N = 3 SE +/- 45.86, N = 9 42359.00 37417.91 2280.18 1. (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 OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 200K 400K 600K 800K 1000K SE +/- 4038.72, N = 3 SE +/- 4724.61, N = 3 SE +/- 4653.04, N = 6 988939.20 611893.07 176350.41 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
ebizzy OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 700K 1400K 2100K 2800K 3500K SE +/- 39888.00, N = 3 SE +/- 8402.66, N = 3 SE +/- 8227.80, N = 15 3075320 2488944 345427 1. (CC) gcc options: -pthread -lpthread -O3 -march=native
TTSIOD 3D Renderer Phong Rendering With Soft-Shadow Mapping OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 200 400 600 800 1000 SE +/- 12.37, N = 5 SE +/- 6.32, N = 3 SE +/- 12.93, N = 12 978.01 881.38 411.62 -msse -mrecip -mfpmath=sse -msse2 -mssse3 -msse -mrecip -mfpmath=sse -msse2 -mssse3 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -lSDL -fopenmp -fwhole-program -lstdc++
Renaissance Test: Genetic Algorithm Using Jenetics + Futures OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Genetic Algorithm Using Jenetics + Futures Graviton2 64c EPYC 7742 64c 128t EPYC 7742 64c 700 1400 2100 2800 3500 SE +/- 10.21, N = 5 SE +/- 41.69, N = 25 SE +/- 17.98, N = 5 2374.70 3143.39 3206.97
Renaissance Test: Akka Unbalanced Cobwebbed Tree OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Akka Unbalanced Cobwebbed Tree EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 6K 12K 18K 24K 30K SE +/- 200.56, N = 5 SE +/- 223.45, N = 5 SE +/- 327.70, N = 25 18076.82 21717.10 26628.12
Renaissance Test: Savina Reactors.IO OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Savina Reactors.IO EPYC 7742 64c EPYC 7742 64c 128t Graviton2 64c 7K 14K 21K 28K 35K SE +/- 184.32, N = 25 SE +/- 297.23, N = 20 SE +/- 446.60, N = 5 16569.16 19664.72 33873.79
Java Gradle Build Gradle Build: Reactor OpenBenchmarking.org Seconds, Fewer Is Better Java Gradle Build Gradle Build: Reactor EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 110 220 330 440 550 SE +/- 11.27, N = 6 SE +/- 2.96, N = 3 SE +/- 18.56, N = 9 340.72 341.38 491.39
Timed MAFFT Alignment Multiple Sequence Alignment OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 7.392 Multiple Sequence Alignment EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 0.5054 1.0108 1.5162 2.0216 2.527 SE +/- 0.029, N = 15 SE +/- 0.013, N = 3 SE +/- 0.049, N = 15 2.136 2.160 2.246 1. (CC) gcc options: -std=c99 -O3 -lm -lpthread
BlogBench Test: Write OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Write EPYC 7742 64c 128t EPYC 7742 64c Graviton2 64c 10K 20K 30K 40K 50K SE +/- 692.73, N = 3 SE +/- 561.49, N = 4 SE +/- 1359.02, N = 3 48646 39604 18017 1. (CC) gcc options: -O2 -pthread
Phoronix Test Suite v10.8.5