EPYC 7F72 2 x AMD EPYC 7F72 24-Core testing with a Supermicro H11DSi-NT v2.00 (2.1 BIOS) and ASPEED on Ubuntu 20.10 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2012196-HA-EPYC7F72759 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 Processor: AMD EPYC 7F72 24-Core @ 3.20GHz (24 Cores / 48 Threads)Motherboard: Supermicro H11DSi-NT v2.00 (2.1 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Western Digital WD_BLACK SN850 1TB, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel 10G X550T
OS: Ubuntu 20.10, Kernel: 5.8.0-29-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.5 Mesa 20.2.1 (LLVM 11.0.0 256 bits), Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080
AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P Processor: 2 x AMD EPYC 7F72 24-Core @ 3.20GHz (48 Cores / 96 Threads)Motherboard: Supermicro H11DSi-NT v2.00 (2.1 BIOS), Chipset: AMD Starship/Matisse, Memory: 126GBDisk: 1000GB Western Digital WD_BLACK SN850 1TB, Graphics: ASPEEDMonitor: VE228, Network: 2 x Intel 10G X550T
OS: Ubuntu 20.10, Kernel: 5.8.0-29-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080
Compiler Notes: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,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 -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301034Python Notes: Python 3.8.6Security Notes: 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 + srbds: Not affected + tsx_async_abort: Not affected
EPYC 7F72 Processor Motherboard Chipset Memory Disk Graphics Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P AMD EPYC 7F72 24-Core @ 3.20GHz (24 Cores / 48 Threads) Supermicro H11DSi-NT v2.00 (2.1 BIOS) AMD Starship/Matisse 64GB 1000GB Western Digital WD_BLACK SN850 1TB llvmpipe VE228 2 x Intel 10G X550T Ubuntu 20.10 5.8.0-29-generic (x86_64) GNOME Shell 3.38.1 X Server 1.20.9 modesetting 1.20.9 4.5 Mesa 20.2.1 (LLVM 11.0.0 256 bits) GCC 10.2.0 ext4 1920x1080 2 x AMD EPYC 7F72 24-Core @ 3.20GHz (48 Cores / 96 Threads) 126GB ASPEED OpenBenchmarking.org Compiler Details - --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,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 - NONE / errors=remount-ro,relatime,rw / Block Size: 4096 Processor Details - Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8301034 Python Details - Python 3.8.6 Security Details - 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 + srbds: Not affected + tsx_async_abort: Not affected
EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P Result Overview Phoronix Test Suite 10.2.2 100% 138% 175% 213% 251% NCNN LevelDB High Performance Conjugate Gradient NAMD BRL-CAD GROMACS Stockfish asmFish LAMMPS Molecular Dynamics Simulator FFTE Timed Linux Kernel Compilation Kvazaar KeyDB Timed LLVM Compilation oneDNN Timed HMMer Search AI Benchmark Alpha Hugin Basis Universal x265 InfluxDB LibRaw Mlpack Benchmark Redis x264 PostgreSQL pgbench HPC Challenge Timed Clash Compilation TNN BYTE Unix Benchmark rav1e LZ4 Compression Numpy Benchmark PHPBench Crafty WebP Image Encode Hierarchical INTegration RNNoise eSpeak-NG Speech Engine TensorFlow Lite
EPYC 7F72 leveldb: Hot Read leveldb: Fill Sync leveldb: Fill Sync leveldb: Overwrite leveldb: Overwrite leveldb: Rand Fill leveldb: Rand Fill leveldb: Rand Read leveldb: Seek Rand leveldb: Rand Delete leveldb: Seq Fill leveldb: Seq Fill yquake2: Software CPU - 1920 x 1080 hpcg: hpcc: G-HPL hpcc: G-Ffte hpcc: EP-DGEMM hpcc: G-Ptrans hpcc: EP-STREAM Triad hpcc: G-Rand Access hpcc: Rand Ring Latency hpcc: Rand Ring Bandwidth hpcc: Max Ping Pong Bandwidth namd: ATPase Simulation - 327,506 Atoms ffte: N=256, 3D Complex FFT Routine hmmer: Pfam Database Search lammps: 20k Atoms lammps: Rhodopsin Protein webp: Default webp: Quality 100 webp: Quality 100, Lossless webp: Quality 100, Highest Compression webp: Quality 100, Lossless, Highest Compression byte: Dhrystone 2 compress-lz4: 1 - Compression Speed compress-lz4: 1 - Decompression Speed compress-lz4: 3 - Compression Speed compress-lz4: 3 - Decompression Speed compress-lz4: 9 - Compression Speed compress-lz4: 9 - Decompression Speed libraw: Post-Processing Benchmark crafty: Elapsed Time onednn: IP Shapes 1D - f32 - CPU onednn: IP Shapes 3D - f32 - CPU onednn: IP Shapes 1D - u8s8f32 - CPU onednn: IP Shapes 3D - u8s8f32 - CPU onednn: Convolution Batch Shapes Auto - f32 - CPU onednn: Deconvolution Batch shapes_1d - f32 - CPU onednn: Deconvolution Batch shapes_3d - f32 - CPU onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU onednn: Recurrent Neural Network Training - f32 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Recurrent Neural Network Training - u8s8f32 - CPU onednn: Recurrent Neural Network Inference - u8s8f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU kvazaar: Bosphorus 4K - Slow kvazaar: Bosphorus 4K - Medium kvazaar: Bosphorus 1080p - Slow kvazaar: Bosphorus 1080p - Medium kvazaar: Bosphorus 4K - Very Fast kvazaar: Bosphorus 4K - Ultra Fast kvazaar: Bosphorus 1080p - Very Fast kvazaar: Bosphorus 1080p - Ultra Fast rav1e: 1 rav1e: 5 rav1e: 6 rav1e: 10 x264: H.264 Video Encoding x265: Bosphorus 4K x265: Bosphorus 1080p stockfish: Total Time asmfish: 1024 Hash Memory, 26 Depth build-clash: Time To Compile build-linux-kernel: Time To Compile build-llvm: Time To Compile numpy: espeak: Text-To-Speech Synthesis rnnoise: keydb: gromacs: Water Benchmark tensorflow-lite: SqueezeNet tensorflow-lite: Inception V4 tensorflow-lite: NASNet Mobile tensorflow-lite: Mobilenet Float tensorflow-lite: Mobilenet Quant tensorflow-lite: Inception ResNet V2 pgbench: 1 - 1 - Read Only pgbench: 1 - 1 - Read Only - Average Latency pgbench: 1 - 1 - Read Write pgbench: 1 - 1 - Read Write - Average Latency pgbench: 1 - 50 - Read Only pgbench: 1 - 50 - Read Only - Average Latency pgbench: 1 - 100 - Read Only pgbench: 1 - 100 - Read Only - Average Latency pgbench: 1 - 50 - Read Write pgbench: 1 - 50 - Read Write - Average Latency pgbench: 100 - 1 - Read Only pgbench: 100 - 1 - Read Only - Average Latency pgbench: 1 - 100 - Read Write pgbench: 1 - 100 - Read Write - Average Latency pgbench: 100 - 1 - Read Write pgbench: 100 - 1 - Read Write - Average Latency pgbench: 100 - 50 - Read Only pgbench: 100 - 50 - Read Only - Average Latency pgbench: 100 - 100 - Read Only pgbench: 100 - 100 - Read Only - Average Latency pgbench: 100 - 50 - Read Write pgbench: 100 - 50 - Read Write - Average Latency pgbench: 100 - 100 - Read Write pgbench: 100 - 100 - Read Write - Average Latency basis: ETC1S basis: UASTC Level 0 basis: UASTC Level 2 basis: UASTC Level 3 basis: UASTC Level 2 + RDO Post-Processing hugin: Panorama Photo Assistant + Stitching Time redis: LPOP redis: SADD redis: LPUSH redis: GET redis: SET ncnn: CPU - squeezenet ncnn: CPU - mobilenet ncnn: CPU-v2-v2 - mobilenet-v2 ncnn: CPU-v3-v3 - mobilenet-v3 ncnn: CPU - shufflenet-v2 ncnn: CPU - mnasnet ncnn: CPU - efficientnet-b0 ncnn: CPU - blazeface ncnn: CPU - googlenet ncnn: CPU - vgg16 ncnn: CPU - resnet18 ncnn: CPU - alexnet ncnn: CPU - resnet50 ncnn: CPU - yolov4-tiny tnn: CPU - MobileNet v2 tnn: CPU - SqueezeNet v1.1 indigobench: CPU - Bedroom indigobench: CPU - Supercar hint: FLOAT ai-benchmark: Device Inference Score ai-benchmark: Device Training Score ai-benchmark: Device AI Score phpbench: PHP Benchmark Suite mlpack: scikit_ica mlpack: scikit_qda mlpack: scikit_svm mlpack: scikit_linearridgeregression brl-cad: VGR Performance Metric influxdb: 4 - 10000 - 2,5000,1 - 10000 influxdb: 64 - 10000 - 2,5000,1 - 10000 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 39.772 4.5 1168.580 23.1 229.121 23.3 228.038 39.867 64.337 209.853 24.1 220.327 14.5 14.9995 87.25450 9.87238 36.14100 7.75739 3.11627 0.03144 1.16869 2.71076 9566.496 0.88169 111181.73685310 142.115 15.789 12.129 1.611 2.596 19.042 8.538 39.280 37455257.2 9777.43 11308.9 50.78 10606.3 49.70 10630.9 35.11 7406309 1.73051 2.77815 1.43782 0.592804 2.84936 2.35946 3.15047 5.30320 6.37725 2.30961 1611.21 925.976 1613.91 934.686 0.578226 1616.57 931.275 1.40235 10.73 10.95 36.21 37.14 24.00 41.99 83.06 142.44 0.349 1.037 1.385 3.039 178.79 23.53 60.61 52597544 64349837 462.283 38.863 298.145 324.13 32.795 21.137 424090.96 2.847 89588.1 1342877 104517 60165.3 61482.0 1175663 29419 0.034 2019 0.495 847231 0.059 698452 0.143 2421 20.663 24336 0.041 2145 46.683 1805 0.554 551623 0.091 529318 0.189 21479 2.329 29486 3.395 49.360 7.578 16.901 27.182 694.869 50.006 2134890.15 1658090.04 1295998.65 2038392.69 1495615.59 21.26 21.44 10.04 9.52 10.45 9.58 12.29 4.09 22.07 36.48 13.82 10.38 24.95 31.05 295.193 275.522 4.769 10.309 321206770.05794 1960 1513 3473 568796 51.74 40.05 24.45 1.61 337924 1197999.8 1339487.5 40.990 4.5 1163.455 23.2 228.697 23.2 228.607 40.383 63.751 210.651 24.0 220.804 23.7 14.9731 87.31077 8.43121 36.76553 8.17553 3.38220 0.03051 1.16276 2.72515 10110.259 0.87532 111004.65873611 142.131 15.816 11.700 1.602 2.585 19.018 8.549 39.170 37547272.9 9740.16 11271.4 50.54 10661.5 48.33 10616.0 35.01 7382916 1.73388 2.75004 1.42645 0.839025 2.90342 2.34152 3.14499 5.35950 6.41390 2.30633 1613.70 930.714 1610.95 930.711 0.575254 1606.75 924.024 1.38651 10.70 10.91 36.09 36.97 23.98 42.13 82.46 141.93 0.349 1.036 1.386 3.032 177.86 23.59 60.33 53421517 64448741 462.323 39.054 289.654 321.81 32.759 21.126 426284.57 2.830 89393.6 1347400 103641 59959.6 61260.7 1179913 29391 0.034 2027 0.493 852612 0.059 698778 0.143 2431 20.571 24345 0.041 2144 46.694 1801 0.556 555320 0.090 530581 0.189 21521 2.324 29537 3.389 49.511 7.733 16.980 27.287 693.897 50.325 1376509.23 1757206.72 1332610.56 1814383.80 1500640.31 21.10 22.26 10.06 9.61 10.35 9.51 12.61 4.13 21.55 36.56 13.86 10.51 24.53 31.33 292.901 275.703 4.775 10.339 321880722.63689 2009 1523 3532 567897 51.82 39.49 24.49 1.59 332394 1199592.4 1339130.1 40.623 4.5 1162.714 23.2 228.708 23.2 228.317 40.030 64.871 210.334 24.0 220.867 14.5 15.4451 86.88297 9.41895 36.43880 8.28555 3.29923 0.03037 1.15717 2.73035 10456.286 0.89277 113220.19747808 142.080 15.779 11.705 1.612 2.591 19.015 8.548 39.342 37624546.9 9780.38 11314.6 49.31 10548.8 48.98 10685.7 34.96 7359217 1.72765 2.78214 1.42669 0.591702 2.83648 2.35726 3.13878 5.36330 6.32884 2.29672 1611.74 929.886 1605.90 928.115 0.576657 1602.59 932.489 1.40346 10.73 10.92 36.15 37.08 23.93 42.17 83.29 142.03 0.346 1.037 1.385 3.026 177.71 23.60 60.25 52704023 63564766 461.624 38.903 294.113 318.47 32.711 21.168 424640.30 2.846 89812.7 1344893 104362 60331.6 61596.2 1178590 29462 0.034 2009 0.498 842022 0.059 695344 0.144 2424 20.630 24446 0.041 2142 46.742 1807 0.553 551484 0.091 529066 0.189 21481 2.328 29552 3.388 49.579 7.606 16.926 27.210 694.995 50.461 1379285.46 1723521.29 1317592.97 1897703.88 1475849.68 21.87 22.20 10.08 9.94 10.45 9.50 12.54 4.16 22.01 37.48 14.10 10.52 25.03 30.75 294.653 275.381 4.770 10.314 322432634.13132 2007 1520 3527 573567 51.84 39.86 25.02 1.62 341885 1197198.0 1297582.5 112.497 6.6 1594.937 14.4 736.377 14.5 731.876 114.896 191.741 690.963 15 707.544 30.3537 100.70033 20.85770 39.20030 15.91823 3.41432 0.04383 2.02418 0.80167 11661.776 0.44651 186579.06690691 187.891 24.924 22.565 1.611 2.573 19.010 8.514 39.013 38124375.5 9657.04 11021.8 49.94 10360.5 48.21 10448.2 31.42 7417621 1.55767 0.784652 2.07247 0.742323 0.860679 2.30564 2.31947 5.49220 2.02580 1.43671 1177.12 823.745 1201.41 838.990 0.518269 1109.29 839.852 0.881752 15.90 16.19 56.46 57.56 34.38 53.26 126.24 181.72 0.348 1.036 1.378 2.972 194.19 20.43 54.70 98232488 115398550 483.404 25.696 209.247 318.08 32.799 21.120 296614.54 5.287 63691.3 871018 117648 40228.6 41754.2 758566 28681 0.035 2016 0.496 816756 0.061 1413247 0.071 2239 22.343 23171 0.043 2020 49.565 1784 0.561 619219 0.081 708234 0.141 23373 2.140 27730 3.612 51.904 7.746 12.391 17.658 687.084 56.785 1388985.48 1818648.35 1365777.40 1888968.34 1498592.43 68.43 46.46 33.18 31.73 31.53 36.45 40.18 11.40 42.95 54.19 21.24 11.07 46.89 45.78 316.823 274.327 322779429.07874 1662 1063 2725 572905 51.72 54.56 24.61 1.78 632495 951085.7 1328030.9 112.695 6.5 1630.936 14.4 736.558 14.5 734.818 112.076 191.127 690.868 14.9 710.333 30.1860 100.37167 20.62957 39.02940 16.45673 3.42379 0.04396 2.05512 0.82649 9370.098 0.44642 185994.09603175 186.396 24.920 22.385 1.604 2.580 18.768 8.523 39.274 37519864.5 9652.00 11114.3 49.67 10409.9 48.83 10473.3 31.14 7419002 1.52395 0.816306 2.03042 0.752290 0.868864 2.26269 2.36572 4.87523 1.99172 1.46841 1166.56 848.249 1182.17 827.223 0.529186 1222.20 822.771 0.890333 15.94 16.13 56.29 57.37 34.31 53.22 126.79 180.39 0.346 1.030 1.369 2.967 193.40 20.20 54.63 97582753 116524954 483.964 25.796 212.024 322.49 32.839 21.080 301799.88 5.278 63239.2 872809 115976 40591.0 41926.2 754906 29257 0.034 2001 0.500 797566 0.063 1407271 0.071 2282 21.917 23295 0.043 2007 49.876 1794 0.558 625512 0.080 715045 0.140 23121 2.164 27720 3.614 51.854 7.808 12.414 17.601 689.471 56.426 1380069.40 1814208.70 1368468.71 1996216.74 1580536.81 39.16 47.44 32.63 32.18 29.73 33.45 47.15 10.84 40.65 55.55 30.48 10.89 44.55 50.92 299.447 274.129 322198292.39955 1695 1078 2773 577796 53.02 54.18 24.51 1.75 638663 949471.8 1345908.1 112.404 6.4 1643.490 14.4 737.227 14.5 733.819 113.380 189.297 670.018 15.1 702.123 28.1161 99.06470 20.07943 35.91387 16.12357 3.21782 0.04279 2.04033 0.80261 8695.062 0.45475 163435.01632332 188.262 24.549 19.962 1.611 2.583 19.198 8.556 39.352 38477706.3 9632.39 11168.7 49.24 10402.7 48.08 10602.4 31.93 7385993 1.67289 0.891776 2.09022 0.792153 0.928138 2.44538 2.30768 3.65062 2.10700 1.50120 1290.07 907.071 1334.58 873.590 0.567497 1407.98 882.055 0.959904 15.83 16.04 56.43 57.31 33.84 52.76 126.88 181.26 0.343 1.016 1.358 2.927 194.97 19.99 53.83 97376595 116189232 482.275 25.950 211.787 320.65 32.829 21.139 308314.45 5.251 76438.6 963296 148263 45501.8 49006.2 823112 29452 0.034 2007 0.498 825448 0.061 1388465 0.072 2276 21.981 23396 0.043 2002 50.006 1766 0.566 633636 0.079 729093 0.137 23360 2.141 27520 3.639 52.049 7.789 12.421 17.704 689.799 58.659 1508834.92 1784823.37 1345722.50 1852789.07 1519952.93 43.06 60.78 34.70 33.50 32.47 39.12 43.86 13.16 57.46 69.56 24.80 12.14 56.48 49.62 322.302 274.450 322561482.00508 1654 1075 2729 572833 53.73 51.46 24.55 1.70 629706 944430.6 1337280.4 OpenBenchmarking.org
High Performance Conjugate Gradient HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 7 14 21 28 35 SE +/- 0.25, N = 12 SE +/- 0.32, N = 12 SE +/- 0.17, N = 3 SE +/- 0.30, N = 3 SE +/- 0.02, N = 3 SE +/- 0.41, N = 12 15.00 14.97 15.45 30.35 30.19 28.12 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi Perf Per Core
OpenBenchmarking.org GFLOP/s Per Core, More Is Better High Performance Conjugate Gradient 3.1 Performance Per Core EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.1448 0.2896 0.4344 0.5792 0.724 0.6250 0.6239 0.6435 0.6324 0.6289 0.5858 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org GFLOP/s Per Thread, More Is Better High Performance Conjugate Gradient 3.1 Performance Per Thread EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0724 0.1448 0.2172 0.2896 0.362 0.3125 0.3119 0.3218 0.3162 0.3144 0.2929 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 7 14 21 28 35 Min: 13.09 / Avg: 15 / Max: 15.67 Min: 12.94 / Avg: 14.97 / Max: 15.76 Min: 15.1 / Avg: 15.45 / Max: 15.65 Min: 30.04 / Avg: 30.35 / Max: 30.94 Min: 30.15 / Avg: 30.19 / Max: 30.22 Min: 25.34 / Avg: 28.12 / Max: 30.13 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
HPC Challenge HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-HPL EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 20 40 60 80 100 SE +/- 0.39, N = 3 SE +/- 0.30, N = 3 SE +/- 0.42, N = 3 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 SE +/- 0.20, N = 3 87.25 87.31 86.88 100.70 100.37 99.06 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org GFLOPS Per Core, More Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: G-HPL EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.819 1.638 2.457 3.276 4.095 3.64 3.64 3.62 2.10 2.09 2.06 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org GFLOPS Per Thread, More Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: G-HPL EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.4095 0.819 1.2285 1.638 2.0475 1.82 1.82 1.81 1.05 1.05 1.03 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-HPL EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 20 40 60 80 100 Min: 86.63 / Avg: 87.25 / Max: 87.98 Min: 86.73 / Avg: 87.31 / Max: 87.73 Min: 86.05 / Avg: 86.88 / Max: 87.42 Min: 100.62 / Avg: 100.7 / Max: 100.75 Min: 100.18 / Avg: 100.37 / Max: 100.52 Min: 98.84 / Avg: 99.06 / Max: 99.45 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
Result
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ffte EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 5 10 15 20 25 SE +/- 1.15527, N = 3 SE +/- 0.56601, N = 3 SE +/- 0.73072, N = 3 SE +/- 0.18996, N = 3 SE +/- 0.25898, N = 3 SE +/- 0.86449, N = 3 9.87238 8.43121 9.41895 20.85770 20.62957 20.07943 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org GFLOPS Per Core, More Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: G-Ffte EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0978 0.1956 0.2934 0.3912 0.489 0.4113 0.3513 0.3925 0.4345 0.4298 0.4183 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org GFLOPS Per Thread, More Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: G-Ffte EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0489 0.0978 0.1467 0.1956 0.2445 0.2057 0.1757 0.1962 0.2173 0.2149 0.2092 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ffte EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 5 10 15 20 25 Min: 7.56 / Avg: 9.87 / Max: 11.05 Min: 7.51 / Avg: 8.43 / Max: 9.46 Min: 8.31 / Avg: 9.42 / Max: 10.8 Min: 20.51 / Avg: 20.86 / Max: 21.16 Min: 20.12 / Avg: 20.63 / Max: 20.96 Min: 18.35 / Avg: 20.08 / Max: 20.97 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
Result
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: EP-DGEMM EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 9 18 27 36 45 SE +/- 0.60, N = 3 SE +/- 0.85, N = 3 SE +/- 0.48, N = 3 SE +/- 0.14, N = 3 SE +/- 0.36, N = 3 SE +/- 0.19, N = 3 36.14 36.77 36.44 39.20 39.03 35.91 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org GFLOPS Per Core, More Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: EP-DGEMM EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.3443 0.6886 1.0329 1.3772 1.7215 1.5100 1.5300 1.5200 0.8167 0.8131 0.7482 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org GFLOPS Per Thread, More Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: EP-DGEMM EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.1723 0.3446 0.5169 0.6892 0.8615 0.7529 0.7659 0.7591 0.4083 0.4066 0.3741 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: EP-DGEMM EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 8 16 24 32 40 Min: 35.08 / Avg: 36.14 / Max: 37.16 Min: 35.57 / Avg: 36.77 / Max: 38.41 Min: 35.49 / Avg: 36.44 / Max: 37.09 Min: 38.93 / Avg: 39.2 / Max: 39.42 Min: 38.32 / Avg: 39.03 / Max: 39.48 Min: 35.53 / Avg: 35.91 / Max: 36.11 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
Result
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ptrans EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 4 8 12 16 20 SE +/- 0.43146, N = 3 SE +/- 0.42362, N = 3 SE +/- 0.32995, N = 3 SE +/- 0.20382, N = 3 SE +/- 0.32220, N = 3 SE +/- 0.83051, N = 3 7.75739 8.17553 8.28555 15.91823 16.45673 16.12357 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org GB/s Per Core, More Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: G-Ptrans EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0777 0.1554 0.2331 0.3108 0.3885 0.3232 0.3406 0.3452 0.3316 0.3428 0.3359 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org GB/s Per Thread, More Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: G-Ptrans EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0388 0.0776 0.1164 0.1552 0.194 0.1616 0.1703 0.1726 0.1658 0.1714 0.1680 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ptrans EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 4 8 12 16 20 Min: 7.29 / Avg: 7.76 / Max: 8.62 Min: 7.33 / Avg: 8.18 / Max: 8.61 Min: 7.63 / Avg: 8.29 / Max: 8.63 Min: 15.64 / Avg: 15.92 / Max: 16.31 Min: 15.92 / Avg: 16.46 / Max: 17.04 Min: 14.46 / Avg: 16.12 / Max: 17.01 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
Result
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: EP-STREAM Triad EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.7704 1.5408 2.3112 3.0816 3.852 SE +/- 0.13922, N = 3 SE +/- 0.00767, N = 3 SE +/- 0.06266, N = 3 SE +/- 0.01458, N = 3 SE +/- 0.00319, N = 3 SE +/- 0.13098, N = 3 3.11627 3.38220 3.29923 3.41432 3.42379 3.21782 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org GB/s Per Core, More Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: EP-STREAM Triad EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0317 0.0634 0.0951 0.1268 0.1585 0.1298 0.1409 0.1375 0.0711 0.0713 0.0670 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org GB/s Per Thread, More Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: EP-STREAM Triad EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0159 0.0318 0.0477 0.0636 0.0795 0.0649 0.0705 0.0687 0.0356 0.0357 0.0335 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: EP-STREAM Triad EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 2.94 / Avg: 3.12 / Max: 3.39 Min: 3.37 / Avg: 3.38 / Max: 3.39 Min: 3.17 / Avg: 3.3 / Max: 3.36 Min: 3.39 / Avg: 3.41 / Max: 3.43 Min: 3.42 / Avg: 3.42 / Max: 3.43 Min: 2.96 / Avg: 3.22 / Max: 3.36 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
Result
OpenBenchmarking.org GUP/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Random Access EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0099 0.0198 0.0297 0.0396 0.0495 SE +/- 0.00015, N = 3 SE +/- 0.00105, N = 3 SE +/- 0.00070, N = 3 SE +/- 0.00017, N = 3 SE +/- 0.00002, N = 3 SE +/- 0.00060, N = 3 0.03144 0.03051 0.03037 0.04383 0.04396 0.04279 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org GUP/s Per Core, More Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: G-Random Access EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0003 0.0006 0.0009 0.0012 0.0015 0.0013 0.0013 0.0013 0.0009 0.0009 0.0009 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org GUP/s Per Thread, More Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: G-Random Access EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0002 0.0004 0.0006 0.0008 0.001 0.0007 0.0006 0.0006 0.0005 0.0005 0.0004 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org GUP/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Random Access EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 1 2 3 4 5 Min: 0.03 / Avg: 0.03 / Max: 0.03 Min: 0.03 / Avg: 0.03 / Max: 0.03 Min: 0.03 / Avg: 0.03 / Max: 0.03 Min: 0.04 / Avg: 0.04 / Max: 0.04 Min: 0.04 / Avg: 0.04 / Max: 0.04 Min: 0.04 / Avg: 0.04 / Max: 0.04 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
Result
OpenBenchmarking.org usecs, Fewer Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Latency EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.4624 0.9248 1.3872 1.8496 2.312 SE +/- 0.00865, N = 3 SE +/- 0.01886, N = 3 SE +/- 0.01464, N = 3 SE +/- 0.01426, N = 3 SE +/- 0.03528, N = 3 SE +/- 0.01822, N = 3 1.16869 1.16276 1.15717 2.02418 2.05512 2.04033 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org usecs x Core, Fewer Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: Random Ring Latency EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 20 40 60 80 100 28.05 27.91 27.77 97.16 98.65 97.94 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org usecs x Thread, Fewer Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: Random Ring Latency EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 40 80 120 160 200 56.10 55.81 55.54 194.32 197.29 195.87 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org usecs, Fewer Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Latency EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 1.16 / Avg: 1.17 / Max: 1.19 Min: 1.13 / Avg: 1.16 / Max: 1.2 Min: 1.13 / Avg: 1.16 / Max: 1.18 Min: 2 / Avg: 2.02 / Max: 2.05 Min: 1.99 / Avg: 2.06 / Max: 2.11 Min: 2.01 / Avg: 2.04 / Max: 2.07 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
Result
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Bandwidth EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.6143 1.2286 1.8429 2.4572 3.0715 SE +/- 0.01911, N = 3 SE +/- 0.06549, N = 3 SE +/- 0.03283, N = 3 SE +/- 0.00576, N = 3 SE +/- 0.02292, N = 3 SE +/- 0.00700, N = 3 2.71076 2.72515 2.73035 0.80167 0.82649 0.80261 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org GB/s Per Core, More Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: Random Ring Bandwidth EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0256 0.0512 0.0768 0.1024 0.128 0.1129 0.1135 0.1138 0.0167 0.0172 0.0167 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org GB/s Per Thread, More Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: Random Ring Bandwidth EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0128 0.0256 0.0384 0.0512 0.064 0.0565 0.0568 0.0569 0.0084 0.0086 0.0084 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Bandwidth EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 2.67 / Avg: 2.71 / Max: 2.74 Min: 2.6 / Avg: 2.73 / Max: 2.81 Min: 2.66 / Avg: 2.73 / Max: 2.76 Min: 0.8 / Avg: 0.8 / Max: 0.81 Min: 0.79 / Avg: 0.83 / Max: 0.87 Min: 0.79 / Avg: 0.8 / Max: 0.81 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
Result
OpenBenchmarking.org MB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Max Ping Pong Bandwidth EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K SE +/- 564.95, N = 3 SE +/- 760.03, N = 3 SE +/- 1135.29, N = 3 SE +/- 491.47, N = 3 SE +/- 437.93, N = 3 SE +/- 296.59, N = 3 9566.50 10110.26 10456.29 11661.78 9370.10 8695.06 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3 Perf Per Core
OpenBenchmarking.org MB/s Per Core, More Is Better HPC Challenge 1.5.0 Performance Per Core - Test / Class: Max Ping Pong Bandwidth EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 90 180 270 360 450 398.60 421.26 435.68 242.95 195.21 181.15 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org MB/s Per Thread, More Is Better HPC Challenge 1.5.0 Performance Per Thread - Test / Class: Max Ping Pong Bandwidth EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 50 100 150 200 250 199.30 210.63 217.84 121.48 97.61 90.57 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org MB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Max Ping Pong Bandwidth EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K Min: 8667.35 / Avg: 9566.5 / Max: 10608.64 Min: 8702.82 / Avg: 10110.26 / Max: 11311.26 Min: 8498.65 / Avg: 10456.29 / Max: 12431.28 Min: 10682.36 / Avg: 11661.78 / Max: 12223.45 Min: 8581.08 / Avg: 9370.1 / Max: 10093.9 Min: 8135.84 / Avg: 8695.06 / Max: 9145.97 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. ATLAS + Open MPI 4.0.3
NAMD NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.2009 0.4018 0.6027 0.8036 1.0045 SE +/- 0.00562, N = 3 SE +/- 0.00402, N = 3 SE +/- 0.01077, N = 3 SE +/- 0.00014, N = 3 SE +/- 0.00027, N = 3 SE +/- 0.00250, N = 3 0.88169 0.87532 0.89277 0.44651 0.44642 0.45475 Perf Per Core
OpenBenchmarking.org days/ns x Core, Fewer Is Better NAMD 2.14 Performance Per Core - ATPase Simulation - 327,506 Atoms EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 5 10 15 20 25 21.16 21.01 21.43 21.43 21.43 21.83 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org days/ns x Thread, Fewer Is Better NAMD 2.14 Performance Per Thread - ATPase Simulation - 327,506 Atoms EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 10 20 30 40 50 42.32 42.02 42.85 42.87 42.86 43.66 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 0.87 / Avg: 0.88 / Max: 0.89 Min: 0.87 / Avg: 0.88 / Max: 0.88 Min: 0.88 / Avg: 0.89 / Max: 0.91 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.45 / Avg: 0.45 / Max: 0.46
FFTE FFTE is a package by Daisuke Takahashi to compute Discrete Fourier Transforms of 1-, 2- and 3- dimensional sequences of length (2^p)*(3^q)*(5^r). Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org MFLOPS, More Is Better FFTE 7.0 N=256, 3D Complex FFT Routine EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 40K 80K 120K 160K 200K SE +/- 1183.09, N = 3 SE +/- 1197.97, N = 4 SE +/- 1183.24, N = 3 SE +/- 2264.92, N = 4 SE +/- 2641.44, N = 3 SE +/- 2147.54, N = 15 111181.74 111004.66 113220.20 186579.07 185994.10 163435.02 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp Perf Per Core
OpenBenchmarking.org MFLOPS Per Core, More Is Better FFTE 7.0 Performance Per Core - N=256, 3D Complex FFT Routine EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 1000 2000 3000 4000 5000 4632.57 4625.19 4717.51 3887.06 3874.88 3404.90 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org MFLOPS Per Thread, More Is Better FFTE 7.0 Performance Per Thread - N=256, 3D Complex FFT Routine EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 500 1000 1500 2000 2500 2316.29 2312.60 2358.75 1943.53 1937.44 1702.45 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org MFLOPS, More Is Better FFTE 7.0 N=256, 3D Complex FFT Routine EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30K 60K 90K 120K 150K Min: 109161.14 / Avg: 111181.74 / Max: 113258.33 Min: 109385.83 / Avg: 111004.66 / Max: 114487.26 Min: 111881.55 / Avg: 113220.2 / Max: 115579.54 Min: 182616.87 / Avg: 186579.07 / Max: 192633.36 Min: 182318.5 / Avg: 185994.1 / Max: 191118.08 Min: 146573.22 / Avg: 163435.02 / Max: 174724.3 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
Timed HMMer Search This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.1 Pfam Database Search EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 40 80 120 160 200 SE +/- 0.11, N = 3 SE +/- 0.02, N = 3 SE +/- 0.14, N = 3 SE +/- 0.64, N = 3 SE +/- 0.78, N = 3 SE +/- 0.09, N = 3 142.12 142.13 142.08 187.89 186.40 188.26 1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm Perf Per Core
OpenBenchmarking.org Seconds x Core, Fewer Is Better Timed HMMer Search 3.3.1 Performance Per Core - Pfam Database Search EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K 3410.76 3411.14 3409.92 9018.77 8947.01 9036.58 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org Seconds x Thread, Fewer Is Better Timed HMMer Search 3.3.1 Performance Per Thread - Pfam Database Search EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 4K 8K 12K 16K 20K 6821.52 6822.29 6819.84 18037.54 17894.02 18073.15 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.1 Pfam Database Search EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 Min: 141.96 / Avg: 142.11 / Max: 142.33 Min: 142.09 / Avg: 142.13 / Max: 142.15 Min: 141.82 / Avg: 142.08 / Max: 142.3 Min: 186.64 / Avg: 187.89 / Max: 188.79 Min: 185.31 / Avg: 186.4 / Max: 187.91 Min: 188.11 / Avg: 188.26 / Max: 188.4 1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
LAMMPS Molecular Dynamics Simulator LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: 20k Atoms EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 15.79 15.82 15.78 24.92 24.92 24.55 1. (CXX) g++ options: -O3 -pthread -lm Perf Per Core
OpenBenchmarking.org ns/day Per Core, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Performance Per Core - Model: 20k Atoms EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.1483 0.2966 0.4449 0.5932 0.7415 0.6579 0.6590 0.6575 0.5193 0.5192 0.5114 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ns/day Per Thread, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Performance Per Thread - Model: 20k Atoms EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0741 0.1482 0.2223 0.2964 0.3705 0.3289 0.3295 0.3287 0.2596 0.2596 0.2557 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: 20k Atoms EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 6 12 18 24 30 Min: 15.75 / Avg: 15.79 / Max: 15.86 Min: 15.71 / Avg: 15.82 / Max: 15.96 Min: 15.76 / Avg: 15.78 / Max: 15.8 Min: 24.88 / Avg: 24.92 / Max: 25 Min: 24.81 / Avg: 24.92 / Max: 25 Min: 24.37 / Avg: 24.55 / Max: 24.68 1. (CXX) g++ options: -O3 -pthread -lm
Result
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: Rhodopsin Protein EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 5 10 15 20 25 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.43, N = 15 SE +/- 0.25, N = 3 SE +/- 0.22, N = 5 12.13 11.70 11.71 22.57 22.39 19.96 1. (CXX) g++ options: -O3 -pthread -lm Perf Per Core
OpenBenchmarking.org ns/day Per Core, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Performance Per Core - Model: Rhodopsin Protein EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.1137 0.2274 0.3411 0.4548 0.5685 0.5054 0.4875 0.4877 0.4701 0.4664 0.4159 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ns/day Per Thread, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Performance Per Thread - Model: Rhodopsin Protein EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.0569 0.1138 0.1707 0.2276 0.2845 0.2527 0.2438 0.2439 0.2351 0.2332 0.2079 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: Rhodopsin Protein EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 5 10 15 20 25 Min: 11.96 / Avg: 12.13 / Max: 12.22 Min: 11.55 / Avg: 11.7 / Max: 11.78 Min: 11.67 / Avg: 11.71 / Max: 11.73 Min: 20.17 / Avg: 22.57 / Max: 24.68 Min: 22.1 / Avg: 22.39 / Max: 22.88 Min: 19.52 / Avg: 19.96 / Max: 20.79 1. (CXX) g++ options: -O3 -pthread -lm
WebP Image Encode This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Default EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.3627 0.7254 1.0881 1.4508 1.8135 SE +/- 0.003, N = 3 SE +/- 0.004, N = 3 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 1.611 1.602 1.612 1.611 1.604 1.611 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff Perf Per Core
OpenBenchmarking.org Encode Time - Seconds x Core, Fewer Is Better WebP Image Encode 1.1 Performance Per Core - Encode Settings: Default EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 20 40 60 80 100 38.66 38.45 38.69 77.33 76.99 77.33 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org Encode Time - Seconds x Thread, Fewer Is Better WebP Image Encode 1.1 Performance Per Thread - Encode Settings: Default EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 77.33 76.90 77.38 154.66 153.98 154.66 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Default EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 1.6 / Avg: 1.61 / Max: 1.62 Min: 1.6 / Avg: 1.6 / Max: 1.61 Min: 1.61 / Avg: 1.61 / Max: 1.62 Min: 1.61 / Avg: 1.61 / Max: 1.61 Min: 1.6 / Avg: 1.6 / Max: 1.61 Min: 1.61 / Avg: 1.61 / Max: 1.61 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.5841 1.1682 1.7523 2.3364 2.9205 SE +/- 0.005, N = 3 SE +/- 0.003, N = 3 SE +/- 0.003, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 SE +/- 0.003, N = 3 2.596 2.585 2.591 2.573 2.580 2.583 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff Perf Per Core
OpenBenchmarking.org Encode Time - Seconds x Core, Fewer Is Better WebP Image Encode 1.1 Performance Per Core - Encode Settings: Quality 100 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 62.30 62.04 62.18 123.50 123.84 123.98 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org Encode Time - Seconds x Thread, Fewer Is Better WebP Image Encode 1.1 Performance Per Thread - Encode Settings: Quality 100 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 50 100 150 200 250 124.61 124.08 124.37 247.01 247.68 247.97 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 2.59 / Avg: 2.6 / Max: 2.6 Min: 2.58 / Avg: 2.58 / Max: 2.59 Min: 2.58 / Avg: 2.59 / Max: 2.6 Min: 2.57 / Avg: 2.57 / Max: 2.57 Min: 2.58 / Avg: 2.58 / Max: 2.58 Min: 2.58 / Avg: 2.58 / Max: 2.59 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.12, N = 3 SE +/- 0.09, N = 3 19.04 19.02 19.02 19.01 18.77 19.20 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff Perf Per Core
OpenBenchmarking.org Encode Time - Seconds x Core, Fewer Is Better WebP Image Encode 1.1 Performance Per Core - Encode Settings: Quality 100, Lossless EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 200 400 600 800 1000 457.01 456.43 456.36 912.48 900.86 921.50 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org Encode Time - Seconds x Thread, Fewer Is Better WebP Image Encode 1.1 Performance Per Thread - Encode Settings: Quality 100, Lossless EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 400 800 1200 1600 2000 914.02 912.86 912.72 1824.96 1801.73 1843.01 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 5 10 15 20 25 Min: 19.02 / Avg: 19.04 / Max: 19.08 Min: 19 / Avg: 19.02 / Max: 19.05 Min: 19.01 / Avg: 19.02 / Max: 19.03 Min: 18.92 / Avg: 19.01 / Max: 19.08 Min: 18.62 / Avg: 18.77 / Max: 19.01 Min: 19.07 / Avg: 19.2 / Max: 19.37 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Highest Compression EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 SE +/- 0.004, N = 3 SE +/- 0.021, N = 3 SE +/- 0.001, N = 3 SE +/- 0.006, N = 3 SE +/- 0.009, N = 3 SE +/- 0.027, N = 3 8.538 8.549 8.548 8.514 8.523 8.556 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff Perf Per Core
OpenBenchmarking.org Encode Time - Seconds x Core, Fewer Is Better WebP Image Encode 1.1 Performance Per Core - Encode Settings: Quality 100, Highest Compression EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 90 180 270 360 450 204.91 205.18 205.15 408.67 409.10 410.69 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org Encode Time - Seconds x Thread, Fewer Is Better WebP Image Encode 1.1 Performance Per Thread - Encode Settings: Quality 100, Highest Compression EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 200 400 600 800 1000 409.82 410.35 410.30 817.34 818.21 821.38 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Highest Compression EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 3 6 9 12 15 Min: 8.53 / Avg: 8.54 / Max: 8.55 Min: 8.53 / Avg: 8.55 / Max: 8.59 Min: 8.55 / Avg: 8.55 / Max: 8.55 Min: 8.51 / Avg: 8.51 / Max: 8.53 Min: 8.51 / Avg: 8.52 / Max: 8.54 Min: 8.53 / Avg: 8.56 / Max: 8.61 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless, Highest Compression EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 9 18 27 36 45 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.13, N = 3 39.28 39.17 39.34 39.01 39.27 39.35 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff Perf Per Core
OpenBenchmarking.org Encode Time - Seconds x Core, Fewer Is Better WebP Image Encode 1.1 Performance Per Core - Encode Settings: Quality 100, Lossless, Highest Compression EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 400 800 1200 1600 2000 942.72 940.08 944.21 1872.62 1885.15 1888.90 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org Encode Time - Seconds x Thread, Fewer Is Better WebP Image Encode 1.1 Performance Per Thread - Encode Settings: Quality 100, Lossless, Highest Compression EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 800 1600 2400 3200 4000 1885.44 1880.16 1888.42 3745.25 3770.30 3777.79 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless, Highest Compression EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 8 16 24 32 40 Min: 39.1 / Avg: 39.28 / Max: 39.38 Min: 39.09 / Avg: 39.17 / Max: 39.33 Min: 39.17 / Avg: 39.34 / Max: 39.49 Min: 38.96 / Avg: 39.01 / Max: 39.11 Min: 39.25 / Avg: 39.27 / Max: 39.32 Min: 39.1 / Avg: 39.35 / Max: 39.57 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
BYTE Unix Benchmark This is a test of BYTE. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 3.6 Computational Test: Dhrystone 2 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 8M 16M 24M 32M 40M SE +/- 295946.74, N = 12 SE +/- 231062.98, N = 3 SE +/- 93255.90, N = 3 SE +/- 218663.85, N = 3 SE +/- 455276.37, N = 3 SE +/- 523722.43, N = 3 37455257.2 37547272.9 37624546.9 38124375.5 37519864.5 38477706.3 Perf Per Core
OpenBenchmarking.org LPS Per Core, More Is Better BYTE Unix Benchmark 3.6 Performance Per Core - Computational Test: Dhrystone 2 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 300K 600K 900K 1200K 1500K 1560635.72 1564469.70 1567689.45 794257.82 781663.84 801618.88 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org LPS Per Thread, More Is Better BYTE Unix Benchmark 3.6 Performance Per Thread - Computational Test: Dhrystone 2 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 200K 400K 600K 800K 1000K 780317.86 782234.85 783844.73 397128.91 390831.92 400809.44 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 3.6 Computational Test: Dhrystone 2 EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 7M 14M 21M 28M 35M Min: 36026099.6 / Avg: 37455257.24 / Max: 39383837.8 Min: 37101722.6 / Avg: 37547272.9 / Max: 37876274.2 Min: 37523480.1 / Avg: 37624546.93 / Max: 37810834.2 Min: 37764703.2 / Avg: 38124375.47 / Max: 38519661.2 Min: 36709387.9 / Avg: 37519864.53 / Max: 38284512.7 Min: 37786879.9 / Avg: 38477706.33 / Max: 39504973.4
LZ4 Compression This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K SE +/- 29.36, N = 3 SE +/- 22.56, N = 3 SE +/- 17.93, N = 3 SE +/- 55.92, N = 3 SE +/- 79.54, N = 3 SE +/- 45.18, N = 3 9777.43 9740.16 9780.38 9657.04 9652.00 9632.39 1. (CC) gcc options: -O3 Perf Per Core
OpenBenchmarking.org MB/s Per Core, More Is Better LZ4 Compression 1.9.3 Performance Per Core - Compression Level: 1 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 90 180 270 360 450 407.39 405.84 407.52 201.19 201.08 200.67 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org MB/s Per Thread, More Is Better LZ4 Compression 1.9.3 Performance Per Thread - Compression Level: 1 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 40 80 120 160 200 203.70 202.92 203.76 100.59 100.54 100.34 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K Min: 9722.63 / Avg: 9777.43 / Max: 9823.08 Min: 9710.44 / Avg: 9740.16 / Max: 9784.42 Min: 9751.48 / Avg: 9780.38 / Max: 9813.23 Min: 9566.8 / Avg: 9657.04 / Max: 9759.38 Min: 9547.98 / Avg: 9652 / Max: 9808.23 Min: 9545.13 / Avg: 9632.39 / Max: 9696.33 1. (CC) gcc options: -O3
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K SE +/- 35.72, N = 3 SE +/- 3.46, N = 3 SE +/- 28.90, N = 3 SE +/- 47.03, N = 3 SE +/- 54.20, N = 3 SE +/- 80.64, N = 3 11308.9 11271.4 11314.6 11021.8 11114.3 11168.7 1. (CC) gcc options: -O3 Perf Per Core
OpenBenchmarking.org MB/s Per Core, More Is Better LZ4 Compression 1.9.3 Performance Per Core - Compression Level: 1 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 100 200 300 400 500 471.20 469.64 471.44 229.62 231.55 232.68 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org MB/s Per Thread, More Is Better LZ4 Compression 1.9.3 Performance Per Thread - Compression Level: 1 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 50 100 150 200 250 235.60 234.82 235.72 114.81 115.77 116.34 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K Min: 11273.1 / Avg: 11308.87 / Max: 11380.3 Min: 11266.3 / Avg: 11271.4 / Max: 11278 Min: 11258.5 / Avg: 11314.6 / Max: 11354.7 Min: 10963.6 / Avg: 11021.8 / Max: 11114.9 Min: 11007 / Avg: 11114.3 / Max: 11181.3 Min: 11014.5 / Avg: 11168.7 / Max: 11286.7 1. (CC) gcc options: -O3
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 11 22 33 44 55 SE +/- 0.55, N = 4 SE +/- 0.36, N = 3 SE +/- 0.05, N = 3 SE +/- 0.49, N = 6 SE +/- 0.61, N = 4 SE +/- 0.56, N = 15 50.78 50.54 49.31 49.94 49.67 49.24 1. (CC) gcc options: -O3 Perf Per Core
OpenBenchmarking.org MB/s Per Core, More Is Better LZ4 Compression 1.9.3 Performance Per Core - Compression Level: 3 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.477 0.954 1.431 1.908 2.385 2.12 2.11 2.05 1.04 1.03 1.03 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org MB/s Per Thread, More Is Better LZ4 Compression 1.9.3 Performance Per Thread - Compression Level: 3 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.2385 0.477 0.7155 0.954 1.1925 1.0600 1.0500 1.0300 0.5202 0.5174 0.5129 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 10 20 30 40 50 Min: 49.99 / Avg: 50.78 / Max: 52.35 Min: 49.82 / Avg: 50.54 / Max: 50.94 Min: 49.22 / Avg: 49.31 / Max: 49.36 Min: 49.12 / Avg: 49.94 / Max: 51.59 Min: 49 / Avg: 49.67 / Max: 51.5 Min: 44.07 / Avg: 49.24 / Max: 51.23 1. (CC) gcc options: -O3
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K SE +/- 37.48, N = 4 SE +/- 1.64, N = 3 SE +/- 18.01, N = 3 SE +/- 35.35, N = 6 SE +/- 72.96, N = 4 SE +/- 35.55, N = 15 10606.3 10661.5 10548.8 10360.5 10409.9 10402.7 1. (CC) gcc options: -O3 Perf Per Core
OpenBenchmarking.org MB/s Per Core, More Is Better LZ4 Compression 1.9.3 Performance Per Core - Compression Level: 3 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 100 200 300 400 500 441.93 444.23 439.53 215.84 216.87 216.72 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org MB/s Per Thread, More Is Better LZ4 Compression 1.9.3 Performance Per Thread - Compression Level: 3 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 50 100 150 200 250 220.96 222.11 219.77 107.92 108.44 108.36 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K Min: 10520 / Avg: 10606.33 / Max: 10683.5 Min: 10658.3 / Avg: 10661.47 / Max: 10663.8 Min: 10513.1 / Avg: 10548.77 / Max: 10571 Min: 10307.4 / Avg: 10360.48 / Max: 10530.7 Min: 10265.2 / Avg: 10409.85 / Max: 10537.8 Min: 10135.6 / Avg: 10402.73 / Max: 10629.1 1. (CC) gcc options: -O3
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 11 22 33 44 55 SE +/- 0.51, N = 5 SE +/- 0.54, N = 5 SE +/- 0.42, N = 3 SE +/- 0.41, N = 3 SE +/- 0.62, N = 3 SE +/- 0.31, N = 3 49.70 48.33 48.98 48.21 48.83 48.08 1. (CC) gcc options: -O3 Perf Per Core
OpenBenchmarking.org MB/s Per Core, More Is Better LZ4 Compression 1.9.3 Performance Per Core - Compression Level: 9 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.4658 0.9316 1.3974 1.8632 2.329 2.07 2.01 2.04 1.00 1.02 1.00 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org MB/s Per Thread, More Is Better LZ4 Compression 1.9.3 Performance Per Thread - Compression Level: 9 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.234 0.468 0.702 0.936 1.17 1.0400 1.0100 1.0200 0.5022 0.5086 0.5008 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Compression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 10 20 30 40 50 Min: 48.53 / Avg: 49.7 / Max: 50.9 Min: 46.91 / Avg: 48.33 / Max: 49.98 Min: 48.14 / Avg: 48.98 / Max: 49.48 Min: 47.75 / Avg: 48.21 / Max: 49.03 Min: 47.77 / Avg: 48.83 / Max: 49.91 Min: 47.46 / Avg: 48.08 / Max: 48.47 1. (CC) gcc options: -O3
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K SE +/- 28.09, N = 5 SE +/- 22.43, N = 5 SE +/- 8.87, N = 3 SE +/- 50.17, N = 3 SE +/- 99.24, N = 3 SE +/- 45.57, N = 3 10630.9 10616.0 10685.7 10448.2 10473.3 10602.4 1. (CC) gcc options: -O3 Perf Per Core
OpenBenchmarking.org MB/s Per Core, More Is Better LZ4 Compression 1.9.3 Performance Per Core - Compression Level: 9 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 100 200 300 400 500 442.95 442.33 445.24 217.67 218.19 220.88 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org MB/s Per Thread, More Is Better LZ4 Compression 1.9.3 Performance Per Thread - Compression Level: 9 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 50 100 150 200 250 221.48 221.17 222.62 108.84 109.10 110.44 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Decompression Speed EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2K 4K 6K 8K 10K Min: 10579.5 / Avg: 10630.88 / Max: 10710.9 Min: 10577.3 / Avg: 10616.04 / Max: 10671 Min: 10668 / Avg: 10685.7 / Max: 10695.7 Min: 10355.3 / Avg: 10448.2 / Max: 10527.5 Min: 10305.6 / Avg: 10473.27 / Max: 10649.1 Min: 10514.3 / Avg: 10602.43 / Max: 10666.6 1. (CC) gcc options: -O3
LibRaw LibRaw is a RAW image decoder for digital camera photos. This test profile runs LibRaw's post-processing benchmark. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org Mpix/sec, More Is Better LibRaw 0.20 Post-Processing Benchmark EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 8 16 24 32 40 SE +/- 0.11, N = 3 SE +/- 0.10, N = 3 SE +/- 0.07, N = 3 SE +/- 0.13, N = 3 SE +/- 0.32, N = 3 SE +/- 0.18, N = 3 35.11 35.01 34.96 31.42 31.14 31.93 1. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm Perf Per Core
OpenBenchmarking.org Mpix/sec Per Core, More Is Better LibRaw 0.20 Performance Per Core - Post-Processing Benchmark EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.3285 0.657 0.9855 1.314 1.6425 1.4600 1.4600 1.4600 0.6546 0.6488 0.6652 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org Mpix/sec Per Thread, More Is Better LibRaw 0.20 Performance Per Thread - Post-Processing Benchmark EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.1646 0.3292 0.4938 0.6584 0.823 0.7315 0.7294 0.7283 0.3273 0.3244 0.3326 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org Mpix/sec, More Is Better LibRaw 0.20 Post-Processing Benchmark EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 8 16 24 32 40 Min: 34.99 / Avg: 35.11 / Max: 35.32 Min: 34.83 / Avg: 35.01 / Max: 35.18 Min: 34.84 / Avg: 34.96 / Max: 35.07 Min: 31.23 / Avg: 31.42 / Max: 31.68 Min: 30.51 / Avg: 31.14 / Max: 31.54 Min: 31.63 / Avg: 31.93 / Max: 32.24 1. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm
Crafty This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page
Result
OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 1.6M 3.2M 4.8M 6.4M 8M SE +/- 14432.27, N = 3 SE +/- 15235.97, N = 3 SE +/- 62574.81, N = 3 SE +/- 1368.48, N = 3 SE +/- 4184.78, N = 3 SE +/- 15876.84, N = 3 7406309 7382916 7359217 7417621 7419002 7385993 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm Perf Per Core
OpenBenchmarking.org Nodes Per Second Per Core, More Is Better Crafty 25.2 Performance Per Core - Elapsed Time EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 70K 140K 210K 280K 350K 308596.21 307621.50 306634.04 154533.77 154562.54 153874.85 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org Nodes Per Second Per Thread, More Is Better Crafty 25.2 Performance Per Thread - Elapsed Time EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30K 60K 90K 120K 150K 154298.10 153810.75 153317.02 77266.89 77281.27 76937.43 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 1.3M 2.6M 3.9M 5.2M 6.5M Min: 7383037 / Avg: 7406309.33 / Max: 7432733 Min: 7361528 / Avg: 7382916.33 / Max: 7412407 Min: 7235678 / Avg: 7359216.67 / Max: 7438320 Min: 7415815 / Avg: 7417621 / Max: 7420305 Min: 7410787 / Avg: 7419001.67 / Max: 7424497 Min: 7354988 / Avg: 7385993.33 / Max: 7407431 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
oneDNN Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.3901 0.7802 1.1703 1.5604 1.9505 SE +/- 0.00365, N = 3 SE +/- 0.01016, N = 3 SE +/- 0.00210, N = 3 SE +/- 0.01626, N = 5 SE +/- 0.01312, N = 3 SE +/- 0.01869, N = 5 1.73051 1.73388 1.72765 1.55767 1.52395 1.67289 MIN: 1.58 MIN: 1.57 MIN: 1.58 MIN: 1.3 MIN: 1.3 MIN: 1.31 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 20 40 60 80 100 41.53 41.61 41.46 74.77 73.15 80.30 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 40 80 120 160 200 83.06 83.23 82.93 149.54 146.30 160.60 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 1.73 / Avg: 1.73 / Max: 1.74 Min: 1.72 / Avg: 1.73 / Max: 1.75 Min: 1.73 / Avg: 1.73 / Max: 1.73 Min: 1.53 / Avg: 1.56 / Max: 1.62 Min: 1.5 / Avg: 1.52 / Max: 1.55 Min: 1.62 / Avg: 1.67 / Max: 1.71 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.626 1.252 1.878 2.504 3.13 SE +/- 0.011681, N = 3 SE +/- 0.017798, N = 3 SE +/- 0.013730, N = 3 SE +/- 0.007882, N = 6 SE +/- 0.009922, N = 3 SE +/- 0.008788, N = 3 2.778150 2.750040 2.782140 0.784652 0.816306 0.891776 MIN: 2.48 MIN: 2.48 MIN: 2.48 MIN: 0.67 MIN: 0.69 MIN: 0.72 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 15 30 45 60 75 66.68 66.00 66.77 37.66 39.18 42.81 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 133.35 132.00 133.54 75.33 78.37 85.61 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 2.76 / Avg: 2.78 / Max: 2.8 Min: 2.73 / Avg: 2.75 / Max: 2.79 Min: 2.76 / Avg: 2.78 / Max: 2.81 Min: 0.76 / Avg: 0.78 / Max: 0.82 Min: 0.8 / Avg: 0.82 / Max: 0.83 Min: 0.87 / Avg: 0.89 / Max: 0.9 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.4703 0.9406 1.4109 1.8812 2.3515 SE +/- 0.00957, N = 3 SE +/- 0.00516, N = 3 SE +/- 0.00056, N = 3 SE +/- 0.02566, N = 4 SE +/- 0.01974, N = 6 SE +/- 0.01869, N = 7 1.43782 1.42645 1.42669 2.07247 2.03042 2.09022 MIN: 1.33 MIN: 1.33 MIN: 1.32 MIN: 1.66 MIN: 1.66 MIN: 1.74 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 20 40 60 80 100 34.51 34.24 34.24 99.48 97.46 100.33 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 40 80 120 160 200 69.02 68.47 68.48 198.96 194.92 200.66 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 1.42 / Avg: 1.44 / Max: 1.45 Min: 1.42 / Avg: 1.43 / Max: 1.44 Min: 1.43 / Avg: 1.43 / Max: 1.43 Min: 2.02 / Avg: 2.07 / Max: 2.14 Min: 1.96 / Avg: 2.03 / Max: 2.08 Min: 2.01 / Avg: 2.09 / Max: 2.14 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.1888 0.3776 0.5664 0.7552 0.944 SE +/- 0.002974, N = 3 SE +/- 0.017578, N = 12 SE +/- 0.000920, N = 3 SE +/- 0.002965, N = 3 SE +/- 0.009511, N = 3 SE +/- 0.001315, N = 3 0.592804 0.839025 0.591702 0.742323 0.752290 0.792153 MIN: 0.51 MIN: 0.64 MIN: 0.5 MIN: 0.67 MIN: 0.67 MIN: 0.66 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 9 18 27 36 45 14.23 20.14 14.20 35.63 36.11 38.02 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 20 40 60 80 100 28.46 40.27 28.40 71.26 72.22 76.05 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 0.59 / Avg: 0.59 / Max: 0.6 Min: 0.74 / Avg: 0.84 / Max: 0.93 Min: 0.59 / Avg: 0.59 / Max: 0.59 Min: 0.74 / Avg: 0.74 / Max: 0.75 Min: 0.73 / Avg: 0.75 / Max: 0.76 Min: 0.79 / Avg: 0.79 / Max: 0.79 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.6533 1.3066 1.9599 2.6132 3.2665 SE +/- 0.008471, N = 3 SE +/- 0.008767, N = 3 SE +/- 0.018414, N = 3 SE +/- 0.011932, N = 3 SE +/- 0.002168, N = 3 SE +/- 0.005045, N = 3 2.849360 2.903420 2.836480 0.860679 0.868864 0.928138 MIN: 2.46 MIN: 2.52 MIN: 2.45 MIN: 0.79 MIN: 0.79 MIN: 0.79 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 16 32 48 64 80 68.39 69.68 68.08 41.31 41.71 44.55 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 136.77 139.36 136.15 82.63 83.41 89.10 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 2.83 / Avg: 2.85 / Max: 2.86 Min: 2.89 / Avg: 2.9 / Max: 2.92 Min: 2.81 / Avg: 2.84 / Max: 2.87 Min: 0.84 / Avg: 0.86 / Max: 0.88 Min: 0.86 / Avg: 0.87 / Max: 0.87 Min: 0.92 / Avg: 0.93 / Max: 0.94 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.5502 1.1004 1.6506 2.2008 2.751 SE +/- 0.00293, N = 3 SE +/- 0.02163, N = 3 SE +/- 0.01986, N = 3 SE +/- 0.03458, N = 12 SE +/- 0.02846, N = 15 SE +/- 0.03614, N = 15 2.35946 2.34152 2.35726 2.30564 2.26269 2.44538 MIN: 2.12 MIN: 2.1 MIN: 2.1 MIN: 1.86 MIN: 1.86 MIN: 1.83 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 56.63 56.20 56.57 110.67 108.61 117.38 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 50 100 150 200 250 113.25 112.39 113.15 221.34 217.22 234.76 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 2.35 / Avg: 2.36 / Max: 2.36 Min: 2.3 / Avg: 2.34 / Max: 2.38 Min: 2.34 / Avg: 2.36 / Max: 2.4 Min: 2.1 / Avg: 2.31 / Max: 2.44 Min: 2.08 / Avg: 2.26 / Max: 2.44 Min: 2.12 / Avg: 2.45 / Max: 2.61 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.7089 1.4178 2.1267 2.8356 3.5445 SE +/- 0.01172, N = 3 SE +/- 0.01746, N = 3 SE +/- 0.01523, N = 3 SE +/- 0.05452, N = 12 SE +/- 0.02839, N = 3 SE +/- 0.03863, N = 15 3.15047 3.14499 3.13878 2.31947 2.36572 2.30768 MIN: 2.98 MIN: 2.97 MIN: 2.98 MIN: 1.96 MIN: 2.01 MIN: 1.95 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 75.61 75.48 75.33 111.34 113.56 110.77 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 50 100 150 200 250 151.22 150.96 150.66 222.67 227.11 221.54 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 3.13 / Avg: 3.15 / Max: 3.17 Min: 3.11 / Avg: 3.14 / Max: 3.16 Min: 3.11 / Avg: 3.14 / Max: 3.17 Min: 2.09 / Avg: 2.32 / Max: 2.7 Min: 2.32 / Avg: 2.37 / Max: 2.42 Min: 2.13 / Avg: 2.31 / Max: 2.57 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 1.2357 2.4714 3.7071 4.9428 6.1785 SE +/- 0.01527, N = 3 SE +/- 0.02370, N = 3 SE +/- 0.03533, N = 3 SE +/- 0.51653, N = 14 SE +/- 0.48144, N = 12 SE +/- 0.04554, N = 3 5.30320 5.35950 5.36330 5.49220 4.87523 3.65062 MIN: 4.97 MIN: 4.96 MIN: 4.96 MIN: 2.85 MIN: 2.89 MIN: 2.99 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 60 120 180 240 300 127.28 128.63 128.72 263.63 234.01 175.23 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 110 220 330 440 550 254.55 257.26 257.44 527.25 468.02 350.46 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 5.27 / Avg: 5.3 / Max: 5.33 Min: 5.31 / Avg: 5.36 / Max: 5.38 Min: 5.3 / Avg: 5.36 / Max: 5.42 Min: 3.26 / Avg: 5.49 / Max: 7.99 Min: 3.2 / Avg: 4.88 / Max: 7.49 Min: 3.59 / Avg: 3.65 / Max: 3.74 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 SE +/- 0.00849, N = 3 SE +/- 0.01141, N = 3 SE +/- 0.03710, N = 3 SE +/- 0.01946, N = 3 SE +/- 0.01945, N = 3 SE +/- 0.01009, N = 3 6.37725 6.41390 6.32884 2.02580 1.99172 2.10700 MIN: 5.79 MIN: 5.78 MIN: 5.76 MIN: 1.87 MIN: 1.86 MIN: 1.86 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 153.05 153.93 151.89 97.24 95.60 101.14 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 70 140 210 280 350 306.11 307.87 303.78 194.48 191.21 202.27 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 3 6 9 12 15 Min: 6.37 / Avg: 6.38 / Max: 6.39 Min: 6.39 / Avg: 6.41 / Max: 6.43 Min: 6.26 / Avg: 6.33 / Max: 6.38 Min: 1.99 / Avg: 2.03 / Max: 2.05 Min: 1.95 / Avg: 1.99 / Max: 2.02 Min: 2.09 / Avg: 2.11 / Max: 2.12 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 0.5197 1.0394 1.5591 2.0788 2.5985 SE +/- 0.00115, N = 3 SE +/- 0.00978, N = 3 SE +/- 0.00103, N = 3 SE +/- 0.00707, N = 3 SE +/- 0.01842, N = 15 SE +/- 0.01121, N = 15 2.30961 2.30633 2.29672 1.43671 1.46841 1.50120 MIN: 2.18 MIN: 2.18 MIN: 2.18 MIN: 1.28 MIN: 1.27 MIN: 1.27 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 16 32 48 64 80 55.43 55.35 55.12 68.96 70.48 72.06 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30 60 90 120 150 110.86 110.70 110.24 137.92 140.97 144.12 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 2 4 6 8 10 Min: 2.31 / Avg: 2.31 / Max: 2.31 Min: 2.29 / Avg: 2.31 / Max: 2.32 Min: 2.29 / Avg: 2.3 / Max: 2.3 Min: 1.42 / Avg: 1.44 / Max: 1.44 Min: 1.39 / Avg: 1.47 / Max: 1.64 Min: 1.46 / Avg: 1.5 / Max: 1.59 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 300 600 900 1200 1500 SE +/- 2.75, N = 3 SE +/- 10.75, N = 3 SE +/- 3.90, N = 3 SE +/- 20.62, N = 14 SE +/- 20.53, N = 15 SE +/- 11.72, N = 3 1611.21 1613.70 1611.74 1177.12 1166.56 1290.07 MIN: 1572.74 MIN: 1565.85 MIN: 1572.94 MIN: 1074.6 MIN: 1069.01 MIN: 1195.72 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread Perf Per Core
OpenBenchmarking.org ms x Core, Fewer Is Better oneDNN 2.0 Performance Per Core - Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 13K 26K 39K 52K 65K 38669.04 38728.80 38681.76 56501.76 55994.88 61923.36 1. EPYC 7F72: Detected core count of 24 2. AMD 7F72: Detected core count of 24 3. AMD EPYC 7F72: Detected core count of 24 4. AMD EPYC 7F72 2P: Detected core count of 48 5. EPYC 7F72 2P: Detected core count of 48 6. 7F72 2P: Detected core count of 48 Perf Per Thread
OpenBenchmarking.org ms x Thread, Fewer Is Better oneDNN 2.0 Performance Per Thread - Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 30K 60K 90K 120K 150K 77338.08 77457.60 77363.52 113003.52 111989.76 123846.72 1. EPYC 7F72: Detected thread count of 48 2. AMD 7F72: Detected thread count of 48 3. AMD EPYC 7F72: Detected thread count of 48 4. AMD EPYC 7F72 2P: Detected thread count of 96 5. EPYC 7F72 2P: Detected thread count of 96 6. 7F72 2P: Detected thread count of 96 Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU EPYC 7F72 AMD 7F72 AMD EPYC 7F72 AMD EPYC 7F72 2P EPYC 7F72 2P 7F72 2P 300 600 900 1200 1500 Min: 1606.72 / Avg: 1611.21 / Max: 1616.2 Min: 1593.97 / Avg: 1613.7 / Max: 1630.96 Min: 1607.18 / Avg: 1611.74 / Max: 1619.5 Min: 1106.82 / Avg: 1177.12 / Max: 1397.87 Min: 1105.59 / Avg: 1166.56 / Max: 1377.85 Min: 1278.32 / Avg: 1290.07 / Max: 1313.51 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread