amazon testing on Ubuntu 22.04 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 2310065-NE-2310055NE35 GCE c3d-standard-60 - Phoronix Test Suite GCE c3d-standard-60 amazon testing on Ubuntu 22.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2310065-NE-2310055NE35&rdt&grs .
GCE c3d-standard-60 Processor Motherboard Chipset Memory Disk Network OS Kernel Vulkan Compiler File-System System Layer c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge AMD EPYC 9B14 (30 Cores / 60 Threads) Google Compute Engine c3d-standard-60 Intel 440FX 82441FX PMC 240GB 215GB nvme_card-pd Google Compute Engine Virtual Ubuntu 22.04 6.2.0-1014-gcp (x86_64) 1.3.238 GCC 11.4.0 ext4 KVM AMD EPYC 7B13 (60 Cores) Google Compute Engine t2d-standard-60 215GB PersistentDisk Red Hat Virtio device ARMv8 Neoverse-N1 (64 Cores) Amazon EC2 c6g.16xlarge (1.0 BIOS) Amazon Device 0200 128GB 215GB Amazon Elastic Block Store Amazon Elastic 5.19.0-1025-aws (aarch64) amazon AMD EPYC 9R14 (64 Cores) Amazon EC2 m7a.16xlarge (1.0 BIOS) Intel 440FX 82441FX PMC 256GB 5.19.0-1025-aws (x86_64) OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - c3d-standard-60 AMD Genoa: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --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-build-config=bootstrap-lto-lean --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 - t2d-standard-60 AMD Milan: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --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-build-config=bootstrap-lto-lean --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 - c6g.16xlarge: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - m7a.16xlarge: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --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-build-config=bootstrap-lto-lean --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 Processor Details - c3d-standard-60 AMD Genoa: CPU Microcode: 0xffffffff - t2d-standard-60 AMD Milan: CPU Microcode: 0xffffffff - m7a.16xlarge: CPU Microcode: 0xa10113e Java Details - OpenJDK Runtime Environment (build 11.0.20.1+1-post-Ubuntu-0ubuntu122.04) Python Details - Python 3.10.12 Security Details - c3d-standard-60 AMD Genoa: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - t2d-standard-60 AMD Milan: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: disabled RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - c6g.16xlarge: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected - m7a.16xlarge: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: disabled RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
GCE c3d-standard-60 openvino: Face Detection FP16-INT8 - CPU openvino: Road Segmentation ADAS FP16-INT8 - CPU openvino: Face Detection Retail FP16-INT8 - CPU npb: BT.C openvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPU openvino: Handwritten English Recognition FP16 - CPU openvino: Road Segmentation ADAS FP16-INT8 - CPU openvino: Handwritten English Recognition FP16-INT8 - CPU openvino: Age Gender Recognition Retail 0013 FP16 - CPU openvino: Person Vehicle Bike Detection FP16 - CPU openvino: Road Segmentation ADAS FP16 - CPU npb: FT.C tensorflow: CPU - 64 - ResNet-50 npb: MG.C avifenc: 2 openssl: ChaCha20-Poly1305 openssl: RSA4096 openssl: ChaCha20 tensorflow: CPU - 32 - ResNet-50 avifenc: 0 openssl: AES-256-GCM openvino: Weld Porosity Detection FP16 - CPU openvino: Vehicle Detection FP16 - CPU openvino: Weld Porosity Detection FP16-INT8 - CPU openvino: Face Detection Retail FP16 - CPU openvino: Face Detection FP16 - CPU tensorflow: CPU - 16 - ResNet-50 openssl: AES-128-GCM build-linux-kernel: defconfig npb: EP.D openvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPU openvino: Vehicle Detection FP16-INT8 - CPU openvino: Person Vehicle Bike Detection FP16 - CPU openvino: Road Segmentation ADAS FP16 - CPU openvino: Machine Translation EN To DE FP16 - CPU openvino: Vehicle Detection FP16 - CPU openvino: Face Detection Retail FP16-INT8 - CPU openvino: Face Detection Retail FP16 - CPU pgbench: 100 - 1000 - Read Only - Average Latency pgbench: 100 - 1000 - Read Only pgbench: 100 - 800 - Read Only pgbench: 100 - 800 - Read Only - Average Latency gromacs: MPI CPU - water_GMX50_bare libxsmm: 32 libxsmm: 64 incompact3d: input.i3d 193 Cells Per Direction remhos: Sample Remap Example incompact3d: input.i3d 129 Cells Per Direction amg: openvino: Face Detection FP16 - CPU openvino: Handwritten English Recognition FP16 - CPU openvino: Vehicle Detection FP16-INT8 - CPU openvino: Face Detection FP16-INT8 - CPU openvino: Handwritten English Recognition FP16-INT8 - CPU openvino: Age Gender Recognition Retail 0013 FP16 - CPU openvino: Machine Translation EN To DE FP16 - CPU openvino: Weld Porosity Detection FP16-INT8 - CPU openssl: RSA4096 openvino: Weld Porosity Detection FP16 - CPU npb: SP.C build-nodejs: Time To Compile openssl: SHA512 openradioss: Chrysler Neon 1M coremark: CoreMark Size 666 - Iterations Per Second avifenc: 6 rodinia: OpenMP CFD Solver lammps: 20k Atoms laghos: Sedov Blast Wave, ube_922_hex.mesh avifenc: 6, Lossless brl-cad: VGR Performance Metric build-linux-kernel: allmodconfig rodinia: OpenMP LavaMD cassandra: Writes openssl: SHA256 build-gem5: Time To Compile heffte: c2c - FFTW - float - 128 nginx: 1000 nginx: 500 compress-7zip: Compression Rating heffte: r2c - FFTW - float - 128 rodinia: OpenMP Leukocyte apache-iotdb: 500 - 100 - 800 - 400 apache-iotdb: 800 - 100 - 800 - 400 blender: Barbershop - CPU-Only apache-iotdb: 800 - 100 - 500 - 400 blender: Classroom - CPU-Only compress-7zip: Decompression Rating laghos: Triple Point Problem blender: BMW27 - CPU-Only apache-iotdb: 500 - 100 - 500 - 400 apache-iotdb: 500 - 100 - 500 - 400 blender: Pabellon Barcelona - CPU-Only blender: Fishy Cat - CPU-Only apache-iotdb: 500 - 100 - 800 - 400 pgbench: 100 - 1000 - Read Write - Average Latency pgbench: 100 - 800 - Read Write - Average Latency openvino: Person Detection FP32 - CPU openvino: Person Detection FP32 - CPU openvino: Person Detection FP16 - CPU openvino: Person Detection FP16 - CPU pgbench: 100 - 1000 - Read Write pgbench: 100 - 800 - Read Write apache-iotdb: 800 - 100 - 800 - 400 apache-iotdb: 800 - 100 - 500 - 400 stockfish: Total Time lammps: Rhodopsin Protein nekrs: TurboPipe Periodic nekrs: Kershaw openradioss: Rubber O-Ring Seal Installation openradioss: Bird Strike on Windshield openradioss: Cell Phone Drop Test openradioss: Bumper Beam heffte: r2c - FFTW - double - 128 heffte: c2c - FFTW - double - 128 rodinia: OpenMP Streamcluster rodinia: OpenMP HotSpot3D npb: LU.C npb: IS.D npb: CG.C c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 340.29 645.74 4.53 96257.48 54971.26 964.46 18.56 761.14 43607.04 1764.21 576.94 39647.47 69.68 42701.83 41.538 123909304773 493077.6 173980949893 62.74 78.068 293328048497 1875.28 1389.69 8.20 4166.39 18.39 50.99 343095284440 3783.60 0.4 2043.05 6.79 20.77 185.29 8.62 6605.12 2.87 4.391 255.4 489.7 28.0196877 33.362 5.87157885 962889833 648.81 31.08 5.86 35.19 39.38 0.52 64.70 3650.57 20079.5 15.98 39919.71 198.394 14702270573 337.70 1445843.521552 3.250 10.025 19.776 259.55 6.889 510819 64.862 228640 46211821313 176.767 88.6301 180537.84 187350.44 271795 148.575 45.498 34762565 35359884 34332237 226211 209.00 418.59 33268158 623.89 83.91 142.90 83.99 142.75 682.12 447.68 105894457 17.423 4723940000 4289858333 89.65 147.31 38.82 92.87 93.6005 57.3116 6.448 84.166 73563.13 2422.40 19597.86 568.77 565.34 3.52 122720.61 44049.15 370.03 26.50 390.72 29668.12 633.76 225.48 54846.18 20.90 47291.96 41.989 119647720337 860844.6 180249145770 20.36 78.350 216025967640 1014.70 368.39 11.32 1285.47 10.73 18.29 234604082610 33.399 4935.68 0.61 1512.57 23.64 66.46 96.58 40.67 4239.52 11.65 0.498 2008186 2003784 0.399 5.289 289.2 554.2 24.5721181 16.326 5.63057327 920427767 1393.56 81.00 9.90 26.28 76.72 0.99 155.14 2646.58 12973.0 14.76 43228.11 191.706 22244804183 327.88 1730658.449440 3.205 7.368 26.734 364.64 7.639 629363 333.351 50.974 187169 50884997103 170.930 109.676 155609.04 162957.75 278973 196.948 42.010 34925899 35068557 351.58 34123810 89.35 247255 222.30 34.27 415.08 33466804 112.64 45.22 633.58 172.717 140.916 193.45 78.96 208.47 73.74 5793 5682 709.46 433.96 112958788 27.828 2730620000 3681935833 72.06 123.61 30.12 75.68 106.029 60.0343 6.423 88.535 94247.77 1752.62 16649.37 22391.86 0.15 2186.81 24229.14 136.16 2.53 6773.31 2.36 178.82 7.36 2.61 21386.37 25661.04 167.946 46715126487 215683.2 67324778360 270.068 129198197600 8.39 6.53 181.94 20.79 0.1 158788510970 102.216 2213.76 7.33 0.14 135.87 382.47 1.36 153.12 0.46 48.08 1.026 975031 1043267 0.767 2.766 312.7 589.5 25.8748328 20.816 5.61811686 1032893667 9996.56 394.94 6990.10 0.04 423.95 5.58 735.58 5.50 2640.0 119.17 9716.99 286.201 14384917863 1259870.716902 4.467 5.983 25.059 321.29 8.879 409.097 62.301 217355 42288513973 224.414 129.172 158700.36 162553.85 239735 202.445 234046 179.52 210.608 168.191 947.86 1.06 947.59 1.06 4776 4784 81807706 26.041 2221710000 1758860000 79.0156 32.3575 14.212 18807.75 915.80 13343.35 261.11 1222.99 3.07 193219.12 92100.80 1419.11 13.07 1158.43 81996.81 3146.04 1049.66 103413.02 100.15 121293.80 35.805 216773475533 996017.5 308308045083 87.19 65.447 522113080527 3132.48 2417.34 5.16 7222.10 31.72 69.55 592545362740 27.709 7501.76 0.27 3666.89 5.07 15.22 315.14 6.60 10382.22 2.21 0.347 2880940 2923009 0.274 7.655 643.4 1201.8 11.5913086 13.867 2.89602661 1843444333 503.38 22.53 4.35 61.19 27.60 0.38 50.72 6177.94 31583.8 10.19 102392.40 154.441 26481506820 190.79 2158639.274883 2.649 6.480 31.471 409.73 5.678 788704 267.965 43.286 278585 62253861197 153.800 121.505 224859.09 233014.72 330633 190.602 34.661 44502333 44699315 276.23 42643903 71.51 282593 218.86 27.74 340.11 40899210 91.88 37.12 521.98 188.676 150.601 56.41 283.40 56.21 284.42 5300 5312 598.49 355.13 135419169 32.785 4774796667 7667846667 59.18 115.96 26.14 66.27 124.363 71.1095 5.930 74.778 210544.87 4085.20 42007.57 OpenBenchmarking.org
OpenVINO Model: Face Detection FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Face Detection FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 5K 10K 15K 20K 25K SE +/- 0.21, N = 3 SE +/- 0.35, N = 3 SE +/- 15.60, N = 3 SE +/- 0.03, N = 3 340.29 568.77 22391.86 261.11 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 22364.17 / MAX: 22423.42 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Road Segmentation ADAS FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Road Segmentation ADAS FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 300 600 900 1200 1500 SE +/- 0.79, N = 3 SE +/- 0.37, N = 3 SE +/- 0.00, N = 3 SE +/- 0.51, N = 3 645.74 565.34 0.15 1222.99 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Face Detection Retail FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Face Detection Retail FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 500 1000 1500 2000 2500 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 27.29, N = 3 SE +/- 0.00, N = 3 4.53 3.52 2186.81 3.07 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 2135.06 / MAX: 2233.34 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
NAS Parallel Benchmarks Test / Class: BT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 40K 80K 120K 160K 200K SE +/- 122.23, N = 3 SE +/- 42.77, N = 3 SE +/- 7.69, N = 3 SE +/- 560.75, N = 3 96257.48 122720.61 24229.14 193219.12 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenVINO Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 20K 40K 60K 80K 100K SE +/- 45.46, N = 3 SE +/- 332.93, N = 3 SE +/- 0.48, N = 3 SE +/- 453.62, N = 3 54971.26 44049.15 136.16 92100.80 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Handwritten English Recognition FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Handwritten English Recognition FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 300 600 900 1200 1500 SE +/- 0.60, N = 3 SE +/- 0.96, N = 3 SE +/- 0.02, N = 3 SE +/- 0.70, N = 3 964.46 370.03 2.53 1419.11 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Road Segmentation ADAS FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Road Segmentation ADAS FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1500 3000 4500 6000 7500 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 76.95, N = 3 SE +/- 0.01, N = 3 18.56 26.50 6773.31 13.07 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 6616.83 / MAX: 6859.99 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Handwritten English Recognition FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Handwritten English Recognition FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 200 400 600 800 1000 SE +/- 1.75, N = 3 SE +/- 1.37, N = 3 SE +/- 0.00, N = 3 SE +/- 0.18, N = 3 761.14 390.72 2.36 1158.43 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 20K 40K 60K 80K 100K SE +/- 18.42, N = 3 SE +/- 14.46, N = 3 SE +/- 0.39, N = 3 SE +/- 31.79, N = 3 43607.04 29668.12 178.82 81996.81 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Person Vehicle Bike Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Person Vehicle Bike Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 700 1400 2100 2800 3500 SE +/- 4.38, N = 3 SE +/- 3.66, N = 3 SE +/- 0.01, N = 3 SE +/- 1.10, N = 3 1764.21 633.76 7.36 3146.04 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Road Segmentation ADAS FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Road Segmentation ADAS FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 200 400 600 800 1000 SE +/- 1.04, N = 3 SE +/- 0.93, N = 3 SE +/- 0.00, N = 3 SE +/- 0.11, N = 3 576.94 225.48 2.61 1049.66 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
NAS Parallel Benchmarks Test / Class: FT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 20K 40K 60K 80K 100K SE +/- 600.19, N = 15 SE +/- 137.77, N = 3 SE +/- 2.85, N = 3 SE +/- 446.85, N = 3 39647.47 54846.18 21386.37 103413.02 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
TensorFlow Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 20 40 60 80 100 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 69.68 20.90 100.15
NAS Parallel Benchmarks Test / Class: MG.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 30K 60K 90K 120K 150K SE +/- 29.69, N = 3 SE +/- 145.06, N = 3 SE +/- 10.99, N = 3 SE +/- 526.14, N = 3 42701.83 47291.96 25661.04 121293.80 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 2 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 40 80 120 160 200 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.22, N = 3 SE +/- 0.22, N = 3 41.54 41.99 167.95 35.81 1. (CXX) g++ options: -O3 -fPIC -lm
OpenSSL Algorithm: ChaCha20-Poly1305 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20-Poly1305 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 50000M 100000M 150000M 200000M 250000M SE +/- 3664727.47, N = 3 SE +/- 198663058.81, N = 3 SE +/- 2259404.37, N = 3 SE +/- 85099636.22, N = 3 123909304773 119647720337 46715126487 216773475533 -m64 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 200K 400K 600K 800K 1000K SE +/- 50.91, N = 3 SE +/- 644.45, N = 3 SE +/- 6.55, N = 3 SE +/- 367.82, N = 3 493077.6 860844.6 215683.2 996017.5 -m64 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenSSL Algorithm: ChaCha20 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 70000M 140000M 210000M 280000M 350000M SE +/- 12326205.97, N = 3 SE +/- 47698640.87, N = 3 SE +/- 372419.81, N = 3 SE +/- 280681661.27, N = 3 173980949893 180249145770 67324778360 308308045083 -m64 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
TensorFlow Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 20 40 60 80 100 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 62.74 20.36 87.19
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 0 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 60 120 180 240 300 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.34, N = 3 SE +/- 0.10, N = 3 78.07 78.35 270.07 65.45 1. (CXX) g++ options: -O3 -fPIC -lm
OpenSSL Algorithm: AES-256-GCM OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-256-GCM c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 110000M 220000M 330000M 440000M 550000M SE +/- 71241287.97, N = 3 SE +/- 178290221.71, N = 3 SE +/- 2100313.05, N = 3 SE +/- 1691817104.57, N = 3 293328048497 216025967640 129198197600 522113080527 -m64 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenVINO Model: Weld Porosity Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Weld Porosity Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 700 1400 2100 2800 3500 SE +/- 0.47, N = 3 SE +/- 0.56, N = 3 SE +/- 0.01, N = 3 SE +/- 0.31, N = 3 1875.28 1014.70 8.39 3132.48 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Vehicle Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Vehicle Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 500 1000 1500 2000 2500 SE +/- 6.33, N = 3 SE +/- 1.55, N = 3 SE +/- 0.01, N = 3 SE +/- 1.65, N = 3 1389.69 368.39 6.53 2417.34 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Weld Porosity Detection FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Weld Porosity Detection FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 40 80 120 160 200 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.24, N = 3 SE +/- 0.01, N = 3 8.20 11.32 181.94 5.16 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 180.71 / MAX: 184.11 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Face Detection Retail FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Face Detection Retail FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1500 3000 4500 6000 7500 SE +/- 8.74, N = 3 SE +/- 15.96, N = 3 SE +/- 0.01, N = 3 SE +/- 4.07, N = 3 4166.39 1285.47 20.79 7222.10 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Face Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Face Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 7 14 21 28 35 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 18.39 10.73 0.10 31.72 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
TensorFlow Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 15 30 45 60 75 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.20, N = 3 50.99 18.29 69.55
OpenSSL Algorithm: AES-128-GCM OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-128-GCM c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 130000M 260000M 390000M 520000M 650000M SE +/- 342949201.09, N = 3 SE +/- 376720190.05, N = 3 SE +/- 5537993.15, N = 3 SE +/- 1818538727.73, N = 3 343095284440 234604082610 158788510970 592545362740 -m64 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
Timed Linux Kernel Compilation Build: defconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: defconfig t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 20 40 60 80 100 SE +/- 0.37, N = 5 SE +/- 0.82, N = 3 SE +/- 0.29, N = 5 33.40 102.22 27.71
NAS Parallel Benchmarks Test / Class: EP.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1600 3200 4800 6400 8000 SE +/- 30.67, N = 3 SE +/- 51.21, N = 5 SE +/- 7.28, N = 3 SE +/- 8.54, N = 3 3783.60 4935.68 2213.76 7501.76 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
OpenVINO Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 0.40 0.61 7.33 0.27 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 6.54 / MAX: 9.95 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Vehicle Detection FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Vehicle Detection FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 800 1600 2400 3200 4000 SE +/- 3.16, N = 3 SE +/- 1.12, N = 3 SE +/- 0.00, N = 15 SE +/- 1.69, N = 3 2043.05 1512.57 0.14 3666.89 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Person Vehicle Bike Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Person Vehicle Bike Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 30 60 90 120 150 SE +/- 0.02, N = 3 SE +/- 0.14, N = 3 SE +/- 0.21, N = 3 SE +/- 0.00, N = 3 6.79 23.64 135.87 5.07 -pie - MIN: 9.57 / MAX: 42.22 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 132.35 / MAX: 148.64 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Road Segmentation ADAS FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Road Segmentation ADAS FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 80 160 240 320 400 SE +/- 0.04, N = 3 SE +/- 0.27, N = 3 SE +/- 0.20, N = 3 SE +/- 0.00, N = 3 20.77 66.46 382.47 15.22 -pie - MIN: 25.85 / MAX: 122.2 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 381.67 / MAX: 383.43 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Machine Translation EN To DE FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Machine Translation EN To DE FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 70 140 210 280 350 SE +/- 0.25, N = 3 SE +/- 0.37, N = 3 SE +/- 0.00, N = 3 SE +/- 0.11, N = 3 185.29 96.58 1.36 315.14 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Vehicle Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Vehicle Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 30 60 90 120 150 SE +/- 0.04, N = 3 SE +/- 0.17, N = 3 SE +/- 0.09, N = 3 SE +/- 0.00, N = 3 8.62 40.67 153.12 6.60 -pie - MIN: 12.07 / MAX: 59.44 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 152.69 / MAX: 153.75 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Face Detection Retail FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Face Detection Retail FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 2K 4K 6K 8K 10K SE +/- 6.52, N = 3 SE +/- 2.08, N = 3 SE +/- 0.01, N = 3 SE +/- 3.60, N = 3 6605.12 4239.52 0.46 10382.22 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Face Detection Retail FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Face Detection Retail FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 11 22 33 44 55 SE +/- 0.01, N = 3 SE +/- 0.14, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 2.87 11.65 48.08 2.21 -pie - MIN: 3.76 / MAX: 29.1 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 47.55 / MAX: 50.49 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 0.2309 0.4618 0.6927 0.9236 1.1545 SE +/- 0.006, N = 3 SE +/- 0.013, N = 3 SE +/- 0.000, N = 3 0.498 1.026 0.347 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 600K 1200K 1800K 2400K 3000K SE +/- 22497.42, N = 3 SE +/- 12606.16, N = 3 SE +/- 2874.04, N = 3 2008186 975031 2880940 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 800 - Mode: Read Only t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 600K 1200K 1800K 2400K 3000K SE +/- 20558.90, N = 3 SE +/- 12058.27, N = 3 SE +/- 13309.30, N = 3 2003784 1043267 2923009 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 0.1726 0.3452 0.5178 0.6904 0.863 SE +/- 0.004, N = 3 SE +/- 0.009, N = 3 SE +/- 0.001, N = 3 0.399 0.767 0.274 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
GROMACS Implementation: MPI CPU - Input: water_GMX50_bare OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2023 Implementation: MPI CPU - Input: water_GMX50_bare c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 2 4 6 8 10 SE +/- 0.011, N = 3 SE +/- 0.005, N = 3 SE +/- 0.001, N = 3 SE +/- 0.035, N = 3 4.391 5.289 2.766 7.655 1. (CXX) g++ options: -O3
libxsmm M N K: 32 OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 32 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 140 280 420 560 700 SE +/- 0.19, N = 3 SE +/- 3.60, N = 4 SE +/- 0.47, N = 3 SE +/- 0.40, N = 3 255.4 289.2 312.7 643.4 -lquadmath -msse4.2 -lquadmath -msse4.2 -march=armv8.1-a -lquadmath -msse4.2 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden
libxsmm M N K: 64 OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 64 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 300 600 900 1200 1500 SE +/- 0.12, N = 3 SE +/- 0.25, N = 3 SE +/- 0.96, N = 3 SE +/- 0.52, N = 3 489.7 554.2 589.5 1201.8 -lquadmath -msse4.2 -lquadmath -msse4.2 -march=armv8.1-a -lquadmath -msse4.2 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden
Xcompact3d Incompact3d Input: input.i3d 193 Cells Per Direction OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 193 Cells Per Direction c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 7 14 21 28 35 SE +/- 0.25, N = 3 SE +/- 0.31, N = 12 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 28.02 24.57 25.87 11.59 1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
Remhos Test: Sample Remap Example OpenBenchmarking.org Seconds, Fewer Is Better Remhos 1.0 Test: Sample Remap Example c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 8 16 24 32 40 SE +/- 0.17, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 33.36 16.33 20.82 13.87 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
Xcompact3d Incompact3d Input: input.i3d 129 Cells Per Direction OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 129 Cells Per Direction c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1.3211 2.6422 3.9633 5.2844 6.6055 SE +/- 0.04970425, N = 3 SE +/- 0.02210564, N = 3 SE +/- 0.01888616, N = 3 SE +/- 0.03993251, N = 3 5.87157885 5.63057327 5.61811686 2.89602661 1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
Algebraic Multi-Grid Benchmark OpenBenchmarking.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 400M 800M 1200M 1600M 2000M SE +/- 2060519.90, N = 3 SE +/- 1088162.98, N = 3 SE +/- 176147.98, N = 3 SE +/- 1428129.35, N = 3 962889833 920427767 1032893667 1843444333 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi
OpenVINO Model: Face Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Face Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 2K 4K 6K 8K 10K SE +/- 0.17, N = 3 SE +/- 1.32, N = 3 SE +/- 1.02, N = 3 SE +/- 0.12, N = 3 648.81 1393.56 9996.56 503.38 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 9993.58 / MAX: 10001.76 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Handwritten English Recognition FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Handwritten English Recognition FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 90 180 270 360 450 SE +/- 0.02, N = 3 SE +/- 0.21, N = 3 SE +/- 2.18, N = 3 SE +/- 0.01, N = 3 31.08 81.00 394.94 22.53 -pie - MIN: 64.65 / MAX: 134.64 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 380.82 / MAX: 408.15 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Vehicle Detection FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Vehicle Detection FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1500 3000 4500 6000 7500 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 27.09, N = 15 SE +/- 0.00, N = 3 5.86 9.90 6990.10 4.35 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 6842.82 / MAX: 7088.53 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Face Detection FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Face Detection FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 14 28 42 56 70 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 35.19 26.28 0.04 61.19 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Handwritten English Recognition FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Handwritten English Recognition FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 90 180 270 360 450 SE +/- 0.09, N = 3 SE +/- 0.27, N = 3 SE +/- 0.26, N = 3 SE +/- 0.01, N = 3 39.38 76.72 423.95 27.60 -pie - MIN: 58.8 / MAX: 121.69 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 411.17 / MAX: 440.57 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1.2555 2.511 3.7665 5.022 6.2775 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 0.52 0.99 5.58 0.38 -pie - MIN: 0.8 / MAX: 13.76 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 5.45 / MAX: 6.48 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Machine Translation EN To DE FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Machine Translation EN To DE FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 160 320 480 640 800 SE +/- 0.09, N = 3 SE +/- 0.58, N = 3 SE +/- 0.26, N = 3 SE +/- 0.02, N = 3 64.70 155.14 735.58 50.72 -pie - MIN: 114.75 / MAX: 224.64 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 734.34 / MAX: 738.23 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Weld Porosity Detection FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Weld Porosity Detection FP16-INT8 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1300 2600 3900 5200 6500 SE +/- 3.28, N = 3 SE +/- 3.26, N = 3 SE +/- 0.01, N = 3 SE +/- 4.43, N = 3 3650.57 2646.58 5.50 6177.94 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 7K 14K 21K 28K 35K SE +/- 14.62, N = 3 SE +/- 11.58, N = 3 SE +/- 0.09, N = 3 SE +/- 24.57, N = 3 20079.5 12973.0 2640.0 31583.8 -m64 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenVINO Model: Weld Porosity Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Weld Porosity Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.00, N = 3 15.98 14.76 119.17 10.19 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 118.73 / MAX: 120.21 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
NAS Parallel Benchmarks Test / Class: SP.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 20K 40K 60K 80K 100K SE +/- 45.01, N = 3 SE +/- 555.92, N = 3 SE +/- 0.76, N = 3 SE +/- 91.17, N = 3 39919.71 43228.11 9716.99 102392.40 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
Timed Node.js Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 19.8.1 Time To Compile c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 60 120 180 240 300 SE +/- 0.29, N = 3 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 198.39 191.71 286.20 154.44
OpenSSL Algorithm: SHA512 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA512 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 6000M 12000M 18000M 24000M 30000M SE +/- 4399663.13, N = 3 SE +/- 108274834.92, N = 3 SE +/- 6214593.12, N = 3 SE +/- 29763937.73, N = 3 14702270573 22244804183 14384917863 26481506820 -m64 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenRadioss Model: Chrysler Neon 1M OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Chrysler Neon 1M c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 70 140 210 280 350 SE +/- 2.03, N = 3 SE +/- 1.48, N = 3 SE +/- 0.61, N = 3 337.70 327.88 190.79
Coremark CoreMark Size 666 - Iterations Per Second OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 500K 1000K 1500K 2000K 2500K SE +/- 1295.68, N = 3 SE +/- 9191.61, N = 3 SE +/- 635.29, N = 3 SE +/- 1437.63, N = 3 1445843.52 1730658.45 1259870.72 2158639.27 1. (CC) gcc options: -O2 -lrt" -lrt
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1.0051 2.0102 3.0153 4.0204 5.0255 SE +/- 0.007, N = 3 SE +/- 0.013, N = 3 SE +/- 0.014, N = 3 SE +/- 0.015, N = 3 3.250 3.205 4.467 2.649 1. (CXX) g++ options: -O3 -fPIC -lm
Rodinia Test: OpenMP CFD Solver OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP CFD Solver c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 3 6 9 12 15 SE +/- 0.013, N = 3 SE +/- 0.034, N = 3 SE +/- 0.001, N = 3 SE +/- 0.007, N = 3 10.025 7.368 5.983 6.480 1. (CXX) g++ options: -O2 -lOpenCL
LAMMPS Molecular Dynamics Simulator Model: 20k Atoms OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 7 14 21 28 35 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 19.78 26.73 25.06 31.47 -lm -lm -lm 1. (CXX) g++ options: -O3 -ldl
Laghos Test: Sedov Blast Wave, ube_922_hex.mesh OpenBenchmarking.org Major Kernels Total Rate, More Is Better Laghos 3.1 Test: Sedov Blast Wave, ube_922_hex.mesh c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 90 180 270 360 450 SE +/- 0.31, N = 3 SE +/- 0.62, N = 3 SE +/- 0.79, N = 3 SE +/- 1.29, N = 3 259.55 364.64 321.29 409.73 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6, Lossless c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 2 4 6 8 10 SE +/- 0.099, N = 3 SE +/- 0.031, N = 3 SE +/- 0.032, N = 3 SE +/- 0.008, N = 3 6.889 7.639 8.879 5.678 1. (CXX) g++ options: -O3 -fPIC -lm
BRL-CAD VGR Performance Metric OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.36 VGR Performance Metric c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 200K 400K 600K 800K 1000K 510819 629363 788704 1. (CXX) g++ options: -std=c++14 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6
Timed Linux Kernel Compilation Build: allmodconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: allmodconfig t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 90 180 270 360 450 SE +/- 1.20, N = 3 SE +/- 2.41, N = 3 SE +/- 0.37, N = 3 333.35 409.10 267.97
Rodinia Test: OpenMP LavaMD OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP LavaMD c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 14 28 42 56 70 SE +/- 0.18, N = 3 SE +/- 0.13, N = 3 SE +/- 0.03, N = 3 SE +/- 0.24, N = 3 64.86 50.97 62.30 43.29 1. (CXX) g++ options: -O2 -lOpenCL
Apache Cassandra Test: Writes OpenBenchmarking.org Op/s, More Is Better Apache Cassandra 4.1.3 Test: Writes c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 60K 120K 180K 240K 300K SE +/- 1129.40, N = 3 SE +/- 681.76, N = 3 SE +/- 3249.94, N = 12 SE +/- 352.65, N = 3 228640 187169 217355 278585
OpenSSL Algorithm: SHA256 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA256 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 13000M 26000M 39000M 52000M 65000M SE +/- 9562123.40, N = 3 SE +/- 20491615.60, N = 3 SE +/- 192235444.29, N = 3 SE +/- 161718701.08, N = 3 46211821313 50884997103 42288513973 62253861197 -m64 -m64 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
Timed Gem5 Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Gem5 Compilation 21.2 Time To Compile c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 50 100 150 200 250 SE +/- 0.07, N = 3 SE +/- 0.27, N = 3 SE +/- 0.03, N = 3 SE +/- 0.39, N = 3 176.77 170.93 224.41 153.80
HeFFTe - Highly Efficient FFT for Exascale Test: c2c - Backend: FFTW - Precision: float - X Y Z: 128 OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: float - X Y Z: 128 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 30 60 90 120 150 SE +/- 0.52, N = 3 SE +/- 0.78, N = 3 SE +/- 0.11, N = 3 SE +/- 1.44, N = 15 88.63 109.68 129.17 121.51 1. (CXX) g++ options: -O3
nginx Connections: 1000 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 50K 100K 150K 200K 250K SE +/- 688.79, N = 3 SE +/- 156.32, N = 3 SE +/- 132.24, N = 3 SE +/- 260.28, N = 3 180537.84 155609.04 158700.36 224859.09 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
nginx Connections: 500 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 50K 100K 150K 200K 250K SE +/- 126.15, N = 3 SE +/- 394.72, N = 3 SE +/- 249.05, N = 3 SE +/- 451.60, N = 3 187350.44 162957.75 162553.85 233014.72 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
7-Zip Compression Test: Compression Rating OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 70K 140K 210K 280K 350K SE +/- 346.51, N = 3 SE +/- 388.74, N = 3 SE +/- 359.95, N = 3 SE +/- 400.99, N = 3 271795 278973 239735 330633 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
HeFFTe - Highly Efficient FFT for Exascale Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128 OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 40 80 120 160 200 SE +/- 2.19, N = 12 SE +/- 0.82, N = 3 SE +/- 0.57, N = 3 SE +/- 2.16, N = 15 148.58 196.95 202.45 190.60 1. (CXX) g++ options: -O3
Rodinia Test: OpenMP Leukocyte OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Leukocyte c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 10 20 30 40 50 SE +/- 0.17, N = 3 SE +/- 0.02, N = 3 SE +/- 0.32, N = 3 45.50 42.01 34.66 1. (CXX) g++ options: -O2 -lOpenCL
Apache IoTDB Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 10M 20M 30M 40M 50M SE +/- 188621.29, N = 3 SE +/- 221111.93, N = 3 SE +/- 224779.59, N = 3 34762565 34925899 44502333
Apache IoTDB Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 10M 20M 30M 40M 50M SE +/- 267152.12, N = 3 SE +/- 354923.23, N = 3 SE +/- 99649.31, N = 3 35359884 35068557 44699315
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Barbershop - Compute: CPU-Only t2d-standard-60 AMD Milan m7a.16xlarge 80 160 240 320 400 SE +/- 0.60, N = 3 SE +/- 0.49, N = 3 351.58 276.23
Apache IoTDB Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 9M 18M 27M 36M 45M SE +/- 66565.66, N = 3 SE +/- 130588.20, N = 3 SE +/- 111073.27, N = 3 34332237 34123810 42643903
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Classroom - Compute: CPU-Only t2d-standard-60 AMD Milan m7a.16xlarge 20 40 60 80 100 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 89.35 71.51
7-Zip Compression Test: Decompression Rating OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 60K 120K 180K 240K 300K SE +/- 519.21, N = 3 SE +/- 347.74, N = 3 SE +/- 57.33, N = 3 SE +/- 342.37, N = 3 226211 247255 234046 282593 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
Laghos Test: Triple Point Problem OpenBenchmarking.org Major Kernels Total Rate, More Is Better Laghos 3.1 Test: Triple Point Problem c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 50 100 150 200 250 SE +/- 0.22, N = 3 SE +/- 1.77, N = 3 SE +/- 0.50, N = 3 SE +/- 1.67, N = 3 209.00 222.30 179.52 218.86 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: BMW27 - Compute: CPU-Only t2d-standard-60 AMD Milan m7a.16xlarge 8 16 24 32 40 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 34.27 27.74
Apache IoTDB Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 90 180 270 360 450 SE +/- 2.14, N = 3 SE +/- 4.08, N = 3 SE +/- 4.92, N = 3 418.59 415.08 340.11 MAX: 31920.67 MAX: 28810.4 MAX: 37899.66
Apache IoTDB Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 9M 18M 27M 36M 45M SE +/- 154587.14, N = 3 SE +/- 303573.79, N = 3 SE +/- 457186.43, N = 3 33268158 33466804 40899210
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Pabellon Barcelona - Compute: CPU-Only t2d-standard-60 AMD Milan m7a.16xlarge 30 60 90 120 150 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 112.64 91.88
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Fishy Cat - Compute: CPU-Only t2d-standard-60 AMD Milan m7a.16xlarge 10 20 30 40 50 SE +/- 0.13, N = 3 SE +/- 0.10, N = 3 45.22 37.12
Apache IoTDB Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 140 280 420 560 700 SE +/- 3.18, N = 3 SE +/- 12.58, N = 3 SE +/- 0.65, N = 3 623.89 633.58 521.98 MAX: 41831.2 MAX: 54749.7 MAX: 34235.44
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 50 100 150 200 250 SE +/- 1.44, N = 8 SE +/- 6.00, N = 9 SE +/- 0.59, N = 3 172.72 210.61 188.68 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 40 80 120 160 200 SE +/- 1.23, N = 12 SE +/- 3.85, N = 12 SE +/- 0.39, N = 3 140.92 168.19 150.60 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenVINO Model: Person Detection FP32 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Person Detection FP32 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 200 400 600 800 1000 SE +/- 0.29, N = 3 SE +/- 7.79, N = 15 SE +/- 0.55, N = 3 SE +/- 0.06, N = 3 83.91 193.45 947.86 56.41 -pie - MIN: 113.44 / MAX: 315.54 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 944.57 / MAX: 958.26 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Person Detection FP32 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Person Detection FP32 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 60 120 180 240 300 SE +/- 0.48, N = 3 SE +/- 2.74, N = 15 SE +/- 0.00, N = 3 SE +/- 0.29, N = 3 142.90 78.96 1.06 283.40 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Person Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.1 Model: Person Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 200 400 600 800 1000 SE +/- 0.13, N = 3 SE +/- 9.17, N = 15 SE +/- 0.37, N = 3 SE +/- 0.10, N = 3 83.99 208.47 947.59 56.21 -pie - MIN: 119.65 / MAX: 316.01 -isystem -fPIC -fvisibility=hidden -std=c++14 -MD -MT -MF -shared - MIN: 943.57 / MAX: 951.68 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
OpenVINO Model: Person Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.1 Model: Person Detection FP16 - Device: CPU c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 60 120 180 240 300 SE +/- 0.23, N = 3 SE +/- 3.12, N = 15 SE +/- 0.00, N = 3 SE +/- 0.53, N = 3 142.75 73.74 1.06 284.42 -pie -pie -pie 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1200 2400 3600 4800 6000 SE +/- 49.93, N = 8 SE +/- 124.50, N = 9 SE +/- 16.57, N = 3 5793 4776 5300 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 800 - Mode: Read Write t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1200 2400 3600 4800 6000 SE +/- 49.28, N = 12 SE +/- 109.40, N = 12 SE +/- 13.63, N = 3 5682 4784 5312 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
Apache IoTDB Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 150 300 450 600 750 SE +/- 7.53, N = 3 SE +/- 25.87, N = 3 SE +/- 4.86, N = 3 682.12 709.46 598.49 MAX: 98294.84 MAX: 113264.06 MAX: 62432.53
Apache IoTDB Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 100 200 300 400 500 SE +/- 27.08, N = 3 SE +/- 35.57, N = 3 SE +/- 7.15, N = 3 447.68 433.96 355.13 MAX: 95136.87 MAX: 103381.73 MAX: 67149.77
Stockfish Total Time OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 15 Total Time c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 30M 60M 90M 120M 150M SE +/- 1450871.09, N = 3 SE +/- 1618403.29, N = 14 SE +/- 1645401.81, N = 15 SE +/- 1001447.21, N = 3 105894457 112958788 81807706 135419169 -m64 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -m64 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -mbmi2 -m64 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 1. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -flto -flto=jobserver
LAMMPS Molecular Dynamics Simulator Model: Rhodopsin Protein OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: Rhodopsin Protein c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 8 16 24 32 40 SE +/- 0.54, N = 12 SE +/- 0.17, N = 3 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 17.42 27.83 26.04 32.79 -lm -lm -lm 1. (CXX) g++ options: -O3 -ldl
nekRS Input: TurboPipe Periodic OpenBenchmarking.org flops/rank, More Is Better nekRS 23.0 Input: TurboPipe Periodic c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1000M 2000M 3000M 4000M 5000M SE +/- 201939132.13, N = 12 SE +/- 481352.26, N = 3 SE +/- 1790009.31, N = 3 SE +/- 6657808.28, N = 3 4723940000 2730620000 2221710000 4774796667 1. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -lmpi
nekRS Input: Kershaw OpenBenchmarking.org flops/rank, More Is Better nekRS 23.0 Input: Kershaw c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 1600M 3200M 4800M 6400M 8000M SE +/- 57202190.49, N = 12 SE +/- 84802173.02, N = 12 SE +/- 2970005.61, N = 3 SE +/- 49077561.86, N = 3 4289858333 3681935833 1758860000 7667846667 1. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -lmpi
OpenRadioss Model: Rubber O-Ring Seal Installation OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Rubber O-Ring Seal Installation c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 20 40 60 80 100 SE +/- 4.01, N = 12 SE +/- 0.22, N = 3 SE +/- 0.73, N = 3 89.65 72.06 59.18
OpenRadioss Model: Bird Strike on Windshield OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Bird Strike on Windshield c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 30 60 90 120 150 SE +/- 3.26, N = 9 SE +/- 0.13, N = 3 SE +/- 0.18, N = 3 147.31 123.61 115.96
OpenRadioss Model: Cell Phone Drop Test OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Cell Phone Drop Test c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 9 18 27 36 45 SE +/- 1.27, N = 15 SE +/- 0.39, N = 15 SE +/- 0.08, N = 3 38.82 30.12 26.14
OpenRadioss Model: Bumper Beam OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Bumper Beam c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 20 40 60 80 100 SE +/- 1.56, N = 15 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 92.87 75.68 66.27
HeFFTe - Highly Efficient FFT for Exascale Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128 OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 30 60 90 120 150 SE +/- 1.78, N = 12 SE +/- 0.91, N = 3 SE +/- 0.71, N = 3 SE +/- 1.18, N = 15 93.60 106.03 79.02 124.36 1. (CXX) g++ options: -O3
HeFFTe - Highly Efficient FFT for Exascale Test: c2c - Backend: FFTW - Precision: double - X Y Z: 128 OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: double - X Y Z: 128 c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 16 32 48 64 80 SE +/- 1.29, N = 15 SE +/- 0.66, N = 3 SE +/- 0.10, N = 3 SE +/- 0.63, N = 15 57.31 60.03 32.36 71.11 1. (CXX) g++ options: -O3
Rodinia Test: OpenMP Streamcluster OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Streamcluster c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 4 8 12 16 20 SE +/- 0.104, N = 15 SE +/- 0.009, N = 3 SE +/- 0.017, N = 3 SE +/- 0.049, N = 3 6.448 6.423 14.212 5.930 1. (CXX) g++ options: -O2 -lOpenCL
Rodinia Test: OpenMP HotSpot3D OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP HotSpot3D c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan m7a.16xlarge 20 40 60 80 100 SE +/- 0.86, N = 15 SE +/- 1.83, N = 12 SE +/- 0.74, N = 15 84.17 88.54 74.78 1. (CXX) g++ options: -O2 -lOpenCL
NAS Parallel Benchmarks Test / Class: LU.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 50K 100K 150K 200K 250K SE +/- 293.57, N = 3 SE +/- 1463.52, N = 15 SE +/- 7.52, N = 3 SE +/- 661.61, N = 3 73563.13 94247.77 18807.75 210544.87 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
NAS Parallel Benchmarks Test / Class: IS.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 900 1800 2700 3600 4500 SE +/- 36.45, N = 15 SE +/- 142.62, N = 12 SE +/- 0.58, N = 3 SE +/- 3.14, N = 3 2422.40 1752.62 915.80 4085.20 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
NAS Parallel Benchmarks Test / Class: CG.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C c3d-standard-60 AMD Genoa t2d-standard-60 AMD Milan c6g.16xlarge m7a.16xlarge 9K 18K 27K 36K 45K SE +/- 77.92, N = 3 SE +/- 1215.82, N = 15 SE +/- 23.52, N = 3 SE +/- 178.46, N = 3 19597.86 16649.37 13343.35 42007.57 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
Phoronix Test Suite v10.8.4