AMD EPYC 7F32 8-Core testing with a ASRockRack EPYCD8 (P2.40 BIOS) and ASPEED on Debian 11 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 2211213-NE-EPTC7F32616 eptc-7f32 - Phoronix Test Suite eptc-7f32 AMD EPYC 7F32 8-Core testing with a ASRockRack EPYCD8 (P2.40 BIOS) and ASPEED on Debian 11 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2211213-NE-EPTC7F32616&sro&grs .
eptc-7f32 Processor Motherboard Chipset Memory Disk Graphics Network OS Kernel Desktop Display Server Compiler File-System Screen Resolution EPYC 7F32 AMD EPYC 7F32 7F32 AMD EPYC 7F32 8-Core @ 3.70GHz (8 Cores / 16 Threads) ASRockRack EPYCD8 (P2.40 BIOS) AMD Starship/Matisse 28GB Samsung SSD 970 EVO Plus 250GB ASPEED 2 x Intel I350 Debian 11 5.10.0-10-amd64 (x86_64) GNOME Shell 3.38.6 X Server GCC 10.2.1 20210110 ext4 1024x768 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: always Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-Km9U7s/gcc-10-10.2.1/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-Km9U7s/gcc-10-10.2.1/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-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 Disk Details - NONE / errors=remount-ro,relatime,rw / Block Size: 4096 Processor Details - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301034 Python Details - Python 3.9.2 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
eptc-7f32 onednn: Convolution Batch Shapes Auto - f32 - CPU onednn: IP Shapes 3D - f32 - CPU openfoam: drivaerFastback, Small Mesh Size - Execution Time mnn: mobilenet-v1-1.0 onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU onednn: Recurrent Neural Network Training - f32 - CPU natron: Spaceship jpegxl-decode: All blosc: blosclz bitshuffle onednn: Recurrent Neural Network Training - u8s8f32 - CPU mnn: MobileNetV2_224 onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU mnn: squeezenetv1.1 onednn: Recurrent Neural Network Inference - u8s8f32 - CPU stress-ng: CPU Cache mnn: SqueezeNetV1.0 tensorflow: CPU - 16 - GoogLeNet openradioss: Cell Phone Drop Test onednn: Recurrent Neural Network Inference - f32 - CPU tensorflow: CPU - 32 - AlexNet tensorflow: CPU - 16 - ResNet-50 onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU tensorflow: CPU - 32 - ResNet-50 onednn: IP Shapes 1D - u8s8f32 - CPU cpuminer-opt: Blake-2 S openfoam: drivaerFastback, Small Mesh Size - Mesh Time tensorflow: CPU - 64 - ResNet-50 aom-av1: Speed 10 Realtime - Bosphorus 1080p cpuminer-opt: x25x mnn: resnet-v2-50 tensorflow: CPU - 32 - GoogLeNet openvino: Person Detection FP32 - CPU y-cruncher: 500M tensorflow: CPU - 64 - AlexNet tensorflow: CPU - 256 - GoogLeNet deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream mnn: mobilenetV3 openvino: Person Detection FP32 - CPU openradioss: INIVOL and Fluid Structure Interaction Drop Container compress-7zip: Compression Rating tensorflow: CPU - 256 - AlexNet aom-av1: Speed 6 Two-Pass - Bosphorus 4K tensorflow: CPU - 16 - AlexNet onednn: Deconvolution Batch shapes_1d - f32 - CPU tensorflow: CPU - 64 - GoogLeNet blosc: blosclz shuffle stress-ng: Context Switching stress-ng: Forking deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream stress-ng: Futex stress-ng: MEMFD openvino: Machine Translation EN To DE FP16 - CPU openvino: Machine Translation EN To DE FP16 - CPU tensorflow: CPU - 512 - AlexNet openvino: Person Vehicle Bike Detection FP16 - CPU openvino: Person Vehicle Bike Detection FP16 - CPU graphics-magick: HWB Color Space deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream avifenc: 10, Lossless jpegxl: JPEG - 80 jpegxl: PNG - 90 stress-ng: Memory Copying stress-ng: CPU Stress openvino: Person Detection FP16 - CPU jpegxl: PNG - 80 xmrig: Monero - 1M aom-av1: Speed 9 Realtime - Bosphorus 4K aom-av1: Speed 10 Realtime - Bosphorus 4K aom-av1: Speed 4 Two-Pass - Bosphorus 4K aom-av1: Speed 9 Realtime - Bosphorus 1080p cpuminer-opt: Ringcoin y-cruncher: 1B jpegxl: JPEG - 90 onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU openvino: Person Detection FP16 - CPU aom-av1: Speed 8 Realtime - Bosphorus 4K smhasher: t1ha0_aes_avx2 x86_64 rocksdb: Read While Writing aom-av1: Speed 8 Realtime - Bosphorus 1080p cpuminer-opt: Magi deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream spacy: en_core_web_trf aom-av1: Speed 6 Realtime - Bosphorus 4K aom-av1: Speed 6 Realtime - Bosphorus 1080p stress-ng: Glibc C String Functions aom-av1: Speed 6 Two-Pass - Bosphorus 1080p openvino: Vehicle Detection FP16 - CPU deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream openvino: Vehicle Detection FP16 - CPU rocksdb: Rand Read stress-ng: MMAP graphics-magick: Resizing aom-av1: Speed 4 Two-Pass - Bosphorus 1080p stress-ng: IO_uring xmrig: Wownero - 1M onednn: IP Shapes 3D - u8s8f32 - CPU jpegxl: PNG - 100 ffmpeg: libx264 - Live ffmpeg: libx264 - Live deepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream encodec: 1.5 kbps avifenc: 6, Lossless build-erlang: Time To Compile nginx: 1000 openvino: Weld Porosity Detection FP16 - CPU openvino: Weld Porosity Detection FP16 - CPU graphics-magick: Rotate onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream scikit-learn: TSNE MNIST Dataset encodec: 6 kbps avifenc: 6 cpuminer-opt: Garlicoin stress-ng: Matrix Math webp2: Default encodec: 3 kbps stress-ng: Socket Activity jpegxl-decode: 1 openradioss: Bird Strike on Windshield stress-ng: Mutex ffmpeg: libx265 - Live nginx: 500 stress-ng: Glibc Qsort Data Sorting ffmpeg: libx265 - Live unpack-linux: linux-5.19.tar.xz deepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Stream deepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Stream avifenc: 0 srsran: 4G PHY_DL_Test 100 PRB SISO 64-QAM openvino: Age Gender Recognition Retail 0013 FP16 - CPU openvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPU spacy: en_core_web_lg tensorflow: CPU - 256 - ResNet-50 deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream openradioss: Rubber O-Ring Seal Installation graphics-magick: Sharpen deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream build-nodejs: Time To Compile build-python: Released Build, PGO + LTO Optimized nginx: 20 openvino: Face Detection FP16-INT8 - CPU deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream openvino: Age Gender Recognition Retail 0013 FP16 - CPU openvino: Weld Porosity Detection FP16-INT8 - CPU webp: Quality 100, Lossless scikit-learn: Sparse Rand Projections, 100 Iterations openvino: Weld Porosity Detection FP16-INT8 - CPU srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAM build-python: Default ffmpeg: libx265 - Platform ffmpeg: libx265 - Platform deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream stress-ng: Crypto deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream nginx: 100 srsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAM cpuminer-opt: LBC, LBRY Credits deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream onednn: Deconvolution Batch shapes_3d - f32 - CPU avifenc: 2 ffmpeg: libx264 - Upload srsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAM stress-ng: Malloc openvino: Vehicle Detection FP16-INT8 - CPU openvino: Vehicle Detection FP16-INT8 - CPU ffmpeg: libx264 - Upload stress-ng: System V Message Passing minibude: OpenMP - BM1 mnn: inception-v3 minibude: OpenMP - BM1 openvino: Face Detection FP16 - CPU graphics-magick: Enhanced openvino: Face Detection FP16-INT8 - CPU blender: BMW27 - CPU-Only smhasher: FarmHash32 x86_64 AVX nginx: 200 openradioss: Bumper Beam ffmpeg: libx265 - Upload srsran: OFDM_Test deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream ffmpeg: libx265 - Upload deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream cpuminer-opt: Skeincoin srsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAM minibude: OpenMP - BM2 minibude: OpenMP - BM2 compress-7zip: Decompression Rating graphics-magick: Noise-Gaussian ffmpeg: libx265 - Video On Demand srsran: 4G PHY_DL_Test 100 PRB SISO 64-QAM ffmpeg: libx265 - Video On Demand stress-ng: SENDFILE onednn: IP Shapes 1D - f32 - CPU srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM encode-flac: WAV To FLAC webp: Quality 100, Highest Compression smhasher: MeowHash x86_64 AES-NI srsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAM webp: Quality 100 cpuminer-opt: scrypt ffmpeg: libx264 - Platform brl-cad: VGR Performance Metric ffmpeg: libx264 - Platform cpuminer-opt: Triple SHA-256, Onecoin build-php: Time To Compile graphics-magick: Swirl stress-ng: Vector Math openvino: Face Detection FP16 - CPU webp: Default srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM scikit-learn: MNIST Dataset openvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPU onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU rocksdb: Update Rand deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream stress-ng: Semaphores deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream encodec: 24 kbps onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU cpuminer-opt: Deepcoin rocksdb: Read Rand Write Rand srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAM smhasher: t1ha2_atonce cpuminer-opt: Myriad-Groestl ffmpeg: libx264 - Video On Demand stress-ng: Atomic stress-ng: NUMA ffmpeg: libx264 - Video On Demand smhasher: SHA3-256 cpuminer-opt: Quad SHA-256, Pyrite smhasher: Spooky32 smhasher: FarmHash128 smhasher: wyhash smhasher: fasthash32 deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream mnn: nasnet aom-av1: Speed 0 Two-Pass - Bosphorus 1080p aom-av1: Speed 0 Two-Pass - Bosphorus 4K webp2: Quality 100, Lossless Compression webp2: Quality 100, Compression Effort 5 webp2: Quality 95, Compression Effort 7 webp2: Quality 75, Compression Effort 7 webp: Quality 100, Lossless, Highest Compression jpegxl: JPEG - 100 smhasher: MeowHash x86_64 AES-NI smhasher: t1ha0_aes_avx2 x86_64 smhasher: FarmHash32 x86_64 AVX smhasher: t1ha2_atonce smhasher: FarmHash128 smhasher: fasthash32 smhasher: Spooky32 smhasher: SHA3-256 smhasher: wyhash EPYC 7F32 AMD EPYC 7F32 7F32 9.55502 8.67144 212.63009 5154.61 5083.34 198.76 5662.8 5251.89 12.2832 1881.32 32.89 130.17 1921.08 65.36 9.98 1864.92 10.22 2.93735 349500 42.034545 11.08 107.35 401.99 32.58 22.892 74.8 34.99 638.29 73076 82.48 9.74 53.09 9.60921 32.75 15057.9 83.98 1184 6.421 8.15 8.26 8.44 7090 41.56 42.08 5.56 97.35 1566.83 45.22 8.03 1.42225 30.22 59489.5 75.1 413.8 20.97 43.34 28.29 1028 12.26 7488.2 0.790351 0.59 28.53 177.00 12.434 120.664 739 1.28 7.893 2139.28 6.29 43.36 330.91 73.77 68.46 7.832 169.538 144.6 12.67 187.5 129 550.805 337.768 1.46 152.3 19.15 253.47 29.88 133.3 20140 6.50887 80.966 227.67 143.3 11.09 270.67 10.827 205 24494.95 173.95 176.3322066 115600000 14.32 68290 353.7 271.092 10.844 58662 271 29.88 328.3 253.56 3.48079 95.1 21.343 3.18 34313.46 328.5 9.90 145.88 42.05 180.15190048 132280 65.041 444 15.80 53.5 3.93432 4.85806 7494.5 353.7 14595.36 12640 41.93 180.64 136.17 66230 13214.7 14311.97 21044.04 6027.75 0.54 0.19 0.01 3.25 0.05 0.12 0.58 0.58 63.122 37.774 47.677 38.503 70.29 40.676 56.219 2843.5 28.387 8.11567 7.35671 184.11155 4.51 4813.42 2.1 246.32 5399.4 5.523 5.78 85.33 9.2 122.99 1891.98 348760 41.187216 102.07 387.05 29.525 1.53 21.633 186.5811 5.3593 3.37 2583.15 627.35 74973 9.89 9.7682 15048.7 5247399 22867.22 5.3725 186.1226 2310980.46 371.54 24 166.48 17.77 224.93 1195 27.1948 36.759 6.206 8.43 8.54 1918.98 18122.93 1.55 8.71 7304 42.81 43.34 5.67 100.17 1612.04 46.211 8.25 2554.02 30.97 59287.09 1565360 76.92 423.63 50.8038 19.6733 618 21.44 42.43 1327261.3 28.46 24.63 6.4357 162.3 39007583 141.82 1046 12.51 4364.73 7636.1 0.59 28.16 179.34 108.5718 36.801 76.2557 13.1116 33.965 12.293 118.859 50516.45 18.79 212.74 739 1.26884 619.198 38.601 34.991 7.856 2123.04 35043.73 6.22 36.202 5664.93 43.72 331.32 5679683.1 74.10 52055.66 134.03 68.15 7.814 37.0454 26.9827 169.306 144.4 5762.94 1.24 11288 70.0202 188.19 130 57.0733 549.975 338.978 46605.09 2.71 6.3581 1.38 29.95 1.46 185.087 267.05 152.5 19.027 252.19 30.04 623.1128 13693.66 80.0283 51130.85 133.4 20260 49.9119 6.54697 80.886 228.158210067 143.3 8923251.96 219.4 18.22 11.07 3150570.51 269.522 38.689 10.781 2.03 206 1468.22 149.11 24384.85 51950.16 174.04 175.58 116100000 158.3902 14.38 25.2198 68130 355.1 270.067 10.803 58704 271 29.97 329.5 252.74 150426.21 3.49229 95.3 21.276 3.18 34206.07 327.5 9.93 146.09 42.14 124747 179.76 132340 65.201 443 39578.06 1957.78 15.83 53.6 119.569 6432.39 352075 54.8712 18.221 1194194.95 22.5985 39.865 7487.49 1196904 354 14583.4 12640 41.96 392425.59 236.71 180.54 136.22 66230 13216.41 14312.8 21040.94 6027.73 176.8383 23.653 0.54 0.19 0.01 3.25 0.05 0.12 0.58 0.58 63.073 37.774 47.677 38.503 70.29 40.684 55.921 2848.539 28.387 13.4819 11.365 269.88197 6.361 6967.79 6477.34 2.8 222.05 4579.4 6350.11 6.602 14.6766 6.713 2184.27 96.63 10.361 29.73 135.8 2084.23 59.61 9.12 2027.82 9.4 2.70369 324270 44.371991 10.36 108.32 410.38 31.285 30.75 1.62 21.766 70.71 33.15 196.8418 5.08 3.201 2456.02 659.67 71377 78.62 9.45 50.79 10.0398 31.39 14437.4 5036770.9 23802.41 5.1633 193.6627 2404426.48 385.73 24.91 160.49 80.98 17.15 232.94 1226 28.1543 35.5064 6.252 8.35 8.52 1983.24 18710.29 1.6 8.63 7299.2 42.42 42.49 5.51 99.91 1575.59 46.501 8.22 1.38568 2490.51 30.82 60752.57 1527948 76.16 422.47 49.6558 20.1274 632 21.33 42.89 1299519.63 27.87 24.12 6.3025 165.72 38208038 144.76 1049 12.34 4451.96 7618.9 0.775961 0.58 28.64 176.34 110.4165 36.2042 77.5095 12.8996 33.419 12.235 119.788 51272.35 18.53 215.66 729 1.26277 627.2384 39.073 34.58 7.802 2147.8 35439.82 6.25 35.801 5728.29 43.24 334.58 5617446.16 73.29 52630.51 132.57 68.90 7.896 37.429 26.7067 167.977 143.3 5811.67 1.23 11197 12.57 69.4795 186.74 130 57.5139 554.104 340.302 46953.57 2.73 6.3119 1.37 29.74 1.45 186.351 268.87 151.5 19.147 253.811141688 29.85 626.9997 13609.36 80.5192 50822.3 132.6 20250 49.6207 6.54145 80.504 228.969381263 142.5 8973004.75 220.62 18.12 11.03 3167586.04 269.315 38.883 10.773 2.04 206 1461.33 148.42 24497.14 52189.35 173.25 176.342980832 115900000 159.0606 14.32 25.1166 68400 354.9 270.907 10.836 58881 272 29.86 328.8 253.642923405 149917.15 3.48771 95.4 21.319 3.19 34310.93 327.7 9.92 145.69 42.03 125073 180.211979224 132610 65.101 444 39495.75 1953.96 15.81 53.6 119.78 6443.59 3.94084 352648 54.949 18.1954 1192572.08 22.5733 39.822 4.86267 7488.34 1195887 354 14594.36 12630 41.94 392155.78 236.87 180.616513283 136.24 66260 13220.33 14307.94 21041.15 6027.12 176.8463 23.653 0.54 0.19 0.01 3.25 0.05 0.12 0.58 0.58 63.111 37.774 47.677 38.503 70.387 40.686 56.139 2840.316 28.387 OpenBenchmarking.org
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 13.48190 8.11567 9.55502 MIN: 13.34 MIN: 7.9 MIN: 9.37 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 11.36500 7.35671 8.67144 MIN: 11.05 MIN: 7.17 MIN: 8.55 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenFOAM Input: drivaerFastback, Small Mesh Size - Execution Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Execution Time 7F32 AMD EPYC 7F32 EPYC 7F32 60 120 180 240 300 269.88 184.11 212.63 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lphysicalProperties -lspecie -lfiniteVolume -lfvModels -lgenericPatchFields -lmeshTools -lsampling -lOpenFOAM -ldl -lm
Mobile Neural Network Model: mobilenet-v1-1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.1 Model: mobilenet-v1-1.0 7F32 AMD EPYC 7F32 2 4 6 8 10 6.361 4.510 MIN: 6.23 / MAX: 14.15 MIN: 4.36 / MAX: 15.87 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU 7F32 EPYC 7F32 1500 3000 4500 6000 7500 6967.79 5154.61 MIN: 6871.03 MIN: 5107.06 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU 7F32 AMD EPYC 7F32 EPYC 7F32 1400 2800 4200 5600 7000 6477.34 4813.42 5083.34 MIN: 6423.47 MIN: 4777.89 MIN: 5017.29 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Natron Input: Spaceship OpenBenchmarking.org FPS, More Is Better Natron 2.4.3 Input: Spaceship 7F32 AMD EPYC 7F32 0.63 1.26 1.89 2.52 3.15 2.8 2.1
JPEG XL Decoding libjxl CPU Threads: All OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.7 CPU Threads: All 7F32 AMD EPYC 7F32 EPYC 7F32 50 100 150 200 250 222.05 246.32 198.76
C-Blosc Test: blosclz bitshuffle OpenBenchmarking.org MB/s, More Is Better C-Blosc 2.3 Test: blosclz bitshuffle 7F32 AMD EPYC 7F32 EPYC 7F32 1200 2400 3600 4800 6000 4579.4 5399.4 5662.8 1. (CC) gcc options: -std=gnu99 -O3 -lrt -pthread -lm
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU 7F32 EPYC 7F32 1400 2800 4200 5600 7000 6350.11 5251.89 MIN: 6292.31 MIN: 5206.92 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.1 Model: MobileNetV2_224 7F32 AMD EPYC 7F32 2 4 6 8 10 6.602 5.523 MIN: 6.53 / MAX: 16.28 MIN: 5.44 / MAX: 6.16 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU 7F32 EPYC 7F32 4 8 12 16 20 14.68 12.28 MIN: 14.24 MIN: 10.92 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Mobile Neural Network Model: squeezenetv1.1 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.1 Model: squeezenetv1.1 7F32 AMD EPYC 7F32 2 4 6 8 10 6.713 5.780 MIN: 6.65 / MAX: 6.91 MIN: 5.73 / MAX: 6.04 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU 7F32 EPYC 7F32 500 1000 1500 2000 2500 2184.27 1881.32 MIN: 2160.57 MIN: 1870.88 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Stress-NG Test: CPU Cache OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: CPU Cache 7F32 AMD EPYC 7F32 20 40 60 80 100 96.63 85.33 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Mobile Neural Network Model: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.1 Model: SqueezeNetV1.0 7F32 AMD EPYC 7F32 3 6 9 12 15 10.361 9.200 MIN: 10 / MAX: 24.43 MIN: 9.09 / MAX: 17.34 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
TensorFlow Device: CPU - Batch Size: 16 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: GoogLeNet 7F32 EPYC 7F32 8 16 24 32 40 29.73 32.89
OpenRadioss Model: Cell Phone Drop Test OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Cell Phone Drop Test 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 135.80 122.99 130.17
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU 7F32 AMD EPYC 7F32 EPYC 7F32 400 800 1200 1600 2000 2084.23 1891.98 1921.08 MIN: 2018.09 MIN: 1850.14 MIN: 1912.63 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
TensorFlow Device: CPU - Batch Size: 32 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: AlexNet 7F32 EPYC 7F32 15 30 45 60 75 59.61 65.36
TensorFlow Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: ResNet-50 7F32 EPYC 7F32 3 6 9 12 15 9.12 9.98
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU 7F32 EPYC 7F32 400 800 1200 1600 2000 2027.82 1864.92 MIN: 1958.42 MIN: 1850.34 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
TensorFlow Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: ResNet-50 7F32 EPYC 7F32 3 6 9 12 15 9.40 10.22
oneDNN Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU 7F32 EPYC 7F32 0.6609 1.3218 1.9827 2.6436 3.3045 2.70369 2.93735 MIN: 2.65 MIN: 2.65 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Cpuminer-Opt Algorithm: Blake-2 S OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Blake-2 S 7F32 AMD EPYC 7F32 EPYC 7F32 70K 140K 210K 280K 350K 324270 348760 349500 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenFOAM Input: drivaerFastback, Small Mesh Size - Mesh Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time 7F32 AMD EPYC 7F32 EPYC 7F32 10 20 30 40 50 44.37 41.19 42.03 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lphysicalProperties -lspecie -lfiniteVolume -lfvModels -lgenericPatchFields -lmeshTools -lsampling -lOpenFOAM -ldl -lm
TensorFlow Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: ResNet-50 7F32 EPYC 7F32 3 6 9 12 15 10.36 11.08
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p 7F32 AMD EPYC 7F32 EPYC 7F32 20 40 60 80 100 108.32 102.07 107.35 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Cpuminer-Opt Algorithm: x25x OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: x25x 7F32 AMD EPYC 7F32 EPYC 7F32 90 180 270 360 450 410.38 387.05 401.99 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Mobile Neural Network Model: resnet-v2-50 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.1 Model: resnet-v2-50 7F32 AMD EPYC 7F32 7 14 21 28 35 31.29 29.53 MIN: 30.43 / MAX: 70.98 MIN: 28.58 / MAX: 56.95 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
TensorFlow Device: CPU - Batch Size: 32 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: GoogLeNet 7F32 EPYC 7F32 8 16 24 32 40 30.75 32.58
OpenVINO Model: Person Detection FP32 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Detection FP32 - Device: CPU 7F32 AMD EPYC 7F32 0.3645 0.729 1.0935 1.458 1.8225 1.62 1.53 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Y-Cruncher Pi Digits To Calculate: 500M OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 500M 7F32 AMD EPYC 7F32 EPYC 7F32 5 10 15 20 25 21.77 21.63 22.89
TensorFlow Device: CPU - Batch Size: 64 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: AlexNet 7F32 EPYC 7F32 20 40 60 80 100 70.71 74.80
TensorFlow Device: CPU - Batch Size: 256 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: GoogLeNet 7F32 EPYC 7F32 8 16 24 32 40 33.15 34.99
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 40 80 120 160 200 196.84 186.58
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 1.2058 2.4116 3.6174 4.8232 6.029 5.0800 5.3593
Mobile Neural Network Model: mobilenetV3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.1 Model: mobilenetV3 7F32 AMD EPYC 7F32 0.7583 1.5166 2.2749 3.0332 3.7915 3.201 3.370 MIN: 3.17 / MAX: 11.24 MIN: 3.34 / MAX: 3.63 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenVINO Model: Person Detection FP32 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Detection FP32 - Device: CPU 7F32 AMD EPYC 7F32 600 1200 1800 2400 3000 2456.02 2583.15 MIN: 2355.49 / MAX: 2547.63 MIN: 2498.05 / MAX: 2737.57 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
OpenRadioss Model: INIVOL and Fluid Structure Interaction Drop Container OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: INIVOL and Fluid Structure Interaction Drop Container 7F32 AMD EPYC 7F32 EPYC 7F32 140 280 420 560 700 659.67 627.35 638.29
7-Zip Compression Test: Compression Rating OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating 7F32 AMD EPYC 7F32 EPYC 7F32 16K 32K 48K 64K 80K 71377 74973 73076 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
TensorFlow Device: CPU - Batch Size: 256 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: AlexNet 7F32 EPYC 7F32 20 40 60 80 100 78.62 82.48
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 9.45 9.89 9.74 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
TensorFlow Device: CPU - Batch Size: 16 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: AlexNet 7F32 EPYC 7F32 12 24 36 48 60 50.79 53.09
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 10.03980 9.76820 9.60921 MIN: 6.19 MIN: 9.06 MIN: 5.99 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
TensorFlow Device: CPU - Batch Size: 64 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: GoogLeNet 7F32 EPYC 7F32 8 16 24 32 40 31.39 32.75
C-Blosc Test: blosclz shuffle OpenBenchmarking.org MB/s, More Is Better C-Blosc 2.3 Test: blosclz shuffle 7F32 AMD EPYC 7F32 EPYC 7F32 3K 6K 9K 12K 15K 14437.4 15048.7 15057.9 1. (CC) gcc options: -std=gnu99 -O3 -lrt -pthread -lm
Stress-NG Test: Context Switching OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Context Switching 7F32 AMD EPYC 7F32 1.1M 2.2M 3.3M 4.4M 5.5M 5036770.9 5247399.0 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Stress-NG Test: Forking OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Forking 7F32 AMD EPYC 7F32 5K 10K 15K 20K 25K 23802.41 22867.22 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 1.2088 2.4176 3.6264 4.8352 6.044 5.1633 5.3725
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 40 80 120 160 200 193.66 186.12
Stress-NG Test: Futex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Futex 7F32 AMD EPYC 7F32 500K 1000K 1500K 2000K 2500K 2404426.48 2310980.46 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Stress-NG Test: MEMFD OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: MEMFD 7F32 AMD EPYC 7F32 80 160 240 320 400 385.73 371.54 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
OpenVINO Model: Machine Translation EN To DE FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU 7F32 AMD EPYC 7F32 6 12 18 24 30 24.91 24.00 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
OpenVINO Model: Machine Translation EN To DE FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU 7F32 AMD EPYC 7F32 40 80 120 160 200 160.49 166.48 MIN: 128.39 / MAX: 175.74 MIN: 141.65 / MAX: 180.81 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
TensorFlow Device: CPU - Batch Size: 512 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 512 - Model: AlexNet 7F32 EPYC 7F32 20 40 60 80 100 80.98 83.98
OpenVINO Model: Person Vehicle Bike Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 4 8 12 16 20 17.15 17.77 MIN: 15.05 / MAX: 24.12 MIN: 13.09 / MAX: 24.14 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
OpenVINO Model: Person Vehicle Bike Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 50 100 150 200 250 232.94 224.93 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: HWB Color Space 7F32 AMD EPYC 7F32 EPYC 7F32 300 600 900 1200 1500 1226 1195 1184 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 7 14 21 28 35 28.15 27.19
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 8 16 24 32 40 35.51 36.76
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 10, Lossless 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 6.252 6.206 6.421 1. (CXX) g++ options: -O3 -fPIC -lm
JPEG XL libjxl Input: JPEG - Quality: 80 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: JPEG - Quality: 80 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 8.35 8.43 8.15 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic
JPEG XL libjxl Input: PNG - Quality: 90 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: PNG - Quality: 90 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 8.52 8.54 8.26 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic
Stress-NG Test: Memory Copying OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Memory Copying 7F32 AMD EPYC 7F32 400 800 1200 1600 2000 1983.24 1918.98 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Stress-NG Test: CPU Stress OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: CPU Stress 7F32 AMD EPYC 7F32 4K 8K 12K 16K 20K 18710.29 18122.93 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
OpenVINO Model: Person Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 0.36 0.72 1.08 1.44 1.8 1.60 1.55 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
JPEG XL libjxl Input: PNG - Quality: 80 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: PNG - Quality: 80 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 8.63 8.71 8.44 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic
Xmrig Variant: Monero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.18.1 Variant: Monero - Hash Count: 1M 7F32 AMD EPYC 7F32 EPYC 7F32 1600 3200 4800 6400 8000 7299.2 7304.0 7090.0 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K 7F32 AMD EPYC 7F32 EPYC 7F32 10 20 30 40 50 42.42 42.81 41.56 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K 7F32 AMD EPYC 7F32 EPYC 7F32 10 20 30 40 50 42.49 43.34 42.08 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K 7F32 AMD EPYC 7F32 EPYC 7F32 1.2758 2.5516 3.8274 5.1032 6.379 5.51 5.67 5.56 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p 7F32 AMD EPYC 7F32 EPYC 7F32 20 40 60 80 100 99.91 100.17 97.35 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Cpuminer-Opt Algorithm: Ringcoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Ringcoin 7F32 AMD EPYC 7F32 EPYC 7F32 300 600 900 1200 1500 1575.59 1612.04 1566.83 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Y-Cruncher Pi Digits To Calculate: 1B OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 1B 7F32 AMD EPYC 7F32 EPYC 7F32 11 22 33 44 55 46.50 46.21 45.22
JPEG XL libjxl Input: JPEG - Quality: 90 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: JPEG - Quality: 90 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 8.22 8.25 8.03 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU 7F32 EPYC 7F32 0.32 0.64 0.96 1.28 1.6 1.38568 1.42225 MIN: 1.3 MIN: 1.3 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenVINO Model: Person Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 500 1000 1500 2000 2500 2490.51 2554.02 MIN: 2401.43 / MAX: 2548.08 MIN: 2208.48 / MAX: 2698.24 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K 7F32 AMD EPYC 7F32 EPYC 7F32 7 14 21 28 35 30.82 30.97 30.22 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
SMHasher Hash: t1ha0_aes_avx2 x86_64 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 7F32 AMD EPYC 7F32 EPYC 7F32 13K 26K 39K 52K 65K 60752.57 59287.09 59489.50 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
Facebook RocksDB Test: Read While Writing OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 7.5.3 Test: Read While Writing 7F32 AMD EPYC 7F32 300K 600K 900K 1200K 1500K 1527948 1565360 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p 7F32 AMD EPYC 7F32 EPYC 7F32 20 40 60 80 100 76.16 76.92 75.10 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Cpuminer-Opt Algorithm: Magi OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Magi 7F32 AMD EPYC 7F32 EPYC 7F32 90 180 270 360 450 422.47 423.63 413.80 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 11 22 33 44 55 49.66 50.80
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 5 10 15 20 25 20.13 19.67
spaCy Model: en_core_web_trf OpenBenchmarking.org tokens/sec, More Is Better spaCy 3.4.1 Model: en_core_web_trf 7F32 AMD EPYC 7F32 140 280 420 560 700 632 618
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K 7F32 AMD EPYC 7F32 EPYC 7F32 5 10 15 20 25 21.33 21.44 20.97 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p 7F32 AMD EPYC 7F32 EPYC 7F32 10 20 30 40 50 42.89 42.43 43.34 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Stress-NG Test: Glibc C String Functions OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Glibc C String Functions 7F32 AMD EPYC 7F32 300K 600K 900K 1200K 1500K 1299519.63 1327261.30 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p 7F32 AMD EPYC 7F32 EPYC 7F32 7 14 21 28 35 27.87 28.46 28.29 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
OpenVINO Model: Vehicle Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 6 12 18 24 30 24.12 24.63 MIN: 20.63 / MAX: 35.91 MIN: 21.3 / MAX: 35.58 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 2 4 6 8 10 6.3025 6.4357
OpenVINO Model: Vehicle Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 40 80 120 160 200 165.72 162.30 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Facebook RocksDB Test: Random Read OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 7.5.3 Test: Random Read 7F32 AMD EPYC 7F32 8M 16M 24M 32M 40M 38208038 39007583 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Stress-NG Test: MMAP OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: MMAP 7F32 AMD EPYC 7F32 30 60 90 120 150 144.76 141.82 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Resizing 7F32 AMD EPYC 7F32 EPYC 7F32 200 400 600 800 1000 1049 1046 1028 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 12.34 12.51 12.26 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Stress-NG Test: IO_uring OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: IO_uring 7F32 AMD EPYC 7F32 1000 2000 3000 4000 5000 4451.96 4364.73 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Xmrig Variant: Wownero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.18.1 Variant: Wownero - Hash Count: 1M 7F32 AMD EPYC 7F32 EPYC 7F32 1600 3200 4800 6400 8000 7618.9 7636.1 7488.2 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU 7F32 EPYC 7F32 0.1778 0.3556 0.5334 0.7112 0.889 0.775961 0.790351 MIN: 0.7 MIN: 0.7 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
JPEG XL libjxl Input: PNG - Quality: 100 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: PNG - Quality: 100 7F32 AMD EPYC 7F32 EPYC 7F32 0.1328 0.2656 0.3984 0.5312 0.664 0.58 0.59 0.59 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic
FFmpeg Encoder: libx264 - Scenario: Live OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Live 7F32 AMD EPYC 7F32 EPYC 7F32 7 14 21 28 35 28.64 28.16 28.53 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx264 - Scenario: Live OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Live 7F32 AMD EPYC 7F32 EPYC 7F32 40 80 120 160 200 176.34 179.34 177.00 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Neural Magic DeepSparse Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 20 40 60 80 100 110.42 108.57
Neural Magic DeepSparse Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 8 16 24 32 40 36.20 36.80
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 20 40 60 80 100 77.51 76.26
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 3 6 9 12 15 12.90 13.11
EnCodec Target Bandwidth: 1.5 kbps OpenBenchmarking.org Seconds, Fewer Is Better EnCodec 0.1.1 Target Bandwidth: 1.5 kbps 7F32 AMD EPYC 7F32 8 16 24 32 40 33.42 33.97
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 6, Lossless 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 12.24 12.29 12.43 1. (CXX) g++ options: -O3 -fPIC -lm
Timed Erlang/OTP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Erlang/OTP Compilation 25.0 Time To Compile 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 119.79 118.86 120.66
nginx Connections: 1000 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 7F32 AMD EPYC 7F32 11K 22K 33K 44K 55K 51272.35 50516.45 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenVINO Model: Weld Porosity Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 5 10 15 20 25 18.53 18.79 MIN: 9.98 / MAX: 29.58 MIN: 10.41 / MAX: 31.29 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
OpenVINO Model: Weld Porosity Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 50 100 150 200 250 215.66 212.74 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Rotate 7F32 AMD EPYC 7F32 EPYC 7F32 160 320 480 640 800 729 739 739 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU 7F32 AMD EPYC 7F32 EPYC 7F32 0.288 0.576 0.864 1.152 1.44 1.26277 1.26884 1.28000 MIN: 1.22 MIN: 1.23 MIN: 1.24 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 140 280 420 560 700 627.24 619.20
Scikit-Learn Benchmark: TSNE MNIST Dataset OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.1.3 Benchmark: TSNE MNIST Dataset 7F32 AMD EPYC 7F32 9 18 27 36 45 39.07 38.60
EnCodec Target Bandwidth: 6 kbps OpenBenchmarking.org Seconds, Fewer Is Better EnCodec 0.1.1 Target Bandwidth: 6 kbps 7F32 AMD EPYC 7F32 8 16 24 32 40 34.58 34.99
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 6 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 7.802 7.856 7.893 1. (CXX) g++ options: -O3 -fPIC -lm
Cpuminer-Opt Algorithm: Garlicoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Garlicoin 7F32 AMD EPYC 7F32 EPYC 7F32 500 1000 1500 2000 2500 2147.80 2123.04 2139.28 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Stress-NG Test: Matrix Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Matrix Math 7F32 AMD EPYC 7F32 8K 16K 24K 32K 40K 35439.82 35043.73 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
WebP2 Image Encode Encode Settings: Default OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Default 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 6.25 6.22 6.29 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl -lpthread
EnCodec Target Bandwidth: 3 kbps OpenBenchmarking.org Seconds, Fewer Is Better EnCodec 0.1.1 Target Bandwidth: 3 kbps 7F32 AMD EPYC 7F32 8 16 24 32 40 35.80 36.20
Stress-NG Test: Socket Activity OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Socket Activity 7F32 AMD EPYC 7F32 1200 2400 3600 4800 6000 5728.29 5664.93 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
JPEG XL Decoding libjxl CPU Threads: 1 OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.7 CPU Threads: 1 7F32 AMD EPYC 7F32 EPYC 7F32 10 20 30 40 50 43.24 43.72 43.36
OpenRadioss Model: Bird Strike on Windshield OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Bird Strike on Windshield 7F32 AMD EPYC 7F32 EPYC 7F32 70 140 210 280 350 334.58 331.32 330.91
Stress-NG Test: Mutex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Mutex 7F32 AMD EPYC 7F32 1.2M 2.4M 3.6M 4.8M 6M 5617446.16 5679683.10 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
FFmpeg Encoder: libx265 - Scenario: Live OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Live 7F32 AMD EPYC 7F32 EPYC 7F32 16 32 48 64 80 73.29 74.10 73.77 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
nginx Connections: 500 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 7F32 AMD EPYC 7F32 11K 22K 33K 44K 55K 52630.51 52055.66 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Stress-NG Test: Glibc Qsort Data Sorting OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Glibc Qsort Data Sorting 7F32 AMD EPYC 7F32 30 60 90 120 150 132.57 134.03 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
FFmpeg Encoder: libx265 - Scenario: Live OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Live 7F32 AMD EPYC 7F32 EPYC 7F32 15 30 45 60 75 68.90 68.15 68.46 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Unpacking The Linux Kernel linux-5.19.tar.xz OpenBenchmarking.org Seconds, Fewer Is Better Unpacking The Linux Kernel 5.19 linux-5.19.tar.xz 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 7.896 7.814 7.832
Neural Magic DeepSparse Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 9 18 27 36 45 37.43 37.05
Neural Magic DeepSparse Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 6 12 18 24 30 26.71 26.98
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 0 7F32 AMD EPYC 7F32 EPYC 7F32 40 80 120 160 200 167.98 169.31 169.54 1. (CXX) g++ options: -O3 -fPIC -lm
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 143.3 144.4 144.6 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
OpenVINO Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU 7F32 AMD EPYC 7F32 1200 2400 3600 4800 6000 5811.67 5762.94 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
OpenVINO Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU 7F32 AMD EPYC 7F32 0.279 0.558 0.837 1.116 1.395 1.23 1.24 MIN: 0.78 / MAX: 8.17 MIN: 0.81 / MAX: 8.45 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
spaCy Model: en_core_web_lg OpenBenchmarking.org tokens/sec, More Is Better spaCy 3.4.1 Model: en_core_web_lg 7F32 AMD EPYC 7F32 2K 4K 6K 8K 10K 11197 11288
TensorFlow Device: CPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: ResNet-50 7F32 EPYC 7F32 3 6 9 12 15 12.57 12.67
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 16 32 48 64 80 69.48 70.02
OpenRadioss Model: Rubber O-Ring Seal Installation OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Rubber O-Ring Seal Installation 7F32 AMD EPYC 7F32 EPYC 7F32 40 80 120 160 200 186.74 188.19 187.50
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Sharpen 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 130 130 129 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 13 26 39 52 65 57.51 57.07
Timed Node.js Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 18.8 Time To Compile 7F32 AMD EPYC 7F32 EPYC 7F32 120 240 360 480 600 554.10 549.98 550.81
Timed CPython Compilation Build Configuration: Released Build, PGO + LTO Optimized OpenBenchmarking.org Seconds, Fewer Is Better Timed CPython Compilation 3.10.6 Build Configuration: Released Build, PGO + LTO Optimized 7F32 AMD EPYC 7F32 EPYC 7F32 70 140 210 280 350 340.30 338.98 337.77
nginx Connections: 20 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 20 7F32 AMD EPYC 7F32 10K 20K 30K 40K 50K 46953.57 46605.09 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenVINO Model: Face Detection FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Face Detection FP16-INT8 - Device: CPU 7F32 AMD EPYC 7F32 0.6143 1.2286 1.8429 2.4572 3.0715 2.73 2.71 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 2 4 6 8 10 6.3119 6.3581
OpenVINO Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU 7F32 AMD EPYC 7F32 0.3105 0.621 0.9315 1.242 1.5525 1.37 1.38 MIN: 0.99 / MAX: 13.32 MIN: 1.02 / MAX: 11.08 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
OpenVINO Model: Weld Porosity Detection FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16-INT8 - Device: CPU 7F32 AMD EPYC 7F32 7 14 21 28 35 29.74 29.95 MIN: 21.45 / MAX: 46.67 MIN: 17.31 / MAX: 48.54 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
WebP Image Encode Encode Settings: Quality 100, Lossless OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless 7F32 AMD EPYC 7F32 EPYC 7F32 0.3285 0.657 0.9855 1.314 1.6425 1.45 1.46 1.46 1. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread
Scikit-Learn Benchmark: Sparse Random Projections, 100 Iterations OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.1.3 Benchmark: Sparse Random Projections, 100 Iterations 7F32 AMD EPYC 7F32 40 80 120 160 200 186.35 185.09
OpenVINO Model: Weld Porosity Detection FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16-INT8 - Device: CPU 7F32 AMD EPYC 7F32 60 120 180 240 300 268.87 267.05 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 151.5 152.5 152.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
Timed CPython Compilation Build Configuration: Default OpenBenchmarking.org Seconds, Fewer Is Better Timed CPython Compilation 3.10.6 Build Configuration: Default 7F32 AMD EPYC 7F32 EPYC 7F32 5 10 15 20 25 19.15 19.03 19.15
FFmpeg Encoder: libx265 - Scenario: Platform OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Platform 7F32 AMD EPYC 7F32 EPYC 7F32 60 120 180 240 300 253.81 252.19 253.47 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx265 - Scenario: Platform OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Platform 7F32 AMD EPYC 7F32 EPYC 7F32 7 14 21 28 35 29.85 30.04 29.88 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 140 280 420 560 700 627.00 623.11
Stress-NG Test: Crypto OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Crypto 7F32 AMD EPYC 7F32 3K 6K 9K 12K 15K 13609.36 13693.66 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 20 40 60 80 100 80.52 80.03
nginx Connections: 100 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 100 7F32 AMD EPYC 7F32 11K 22K 33K 44K 55K 50822.30 51130.85 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 132.6 133.4 133.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
Cpuminer-Opt Algorithm: LBC, LBRY Credits OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: LBC, LBRY Credits 7F32 AMD EPYC 7F32 EPYC 7F32 4K 8K 12K 16K 20K 20250 20260 20140 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 11 22 33 44 55 49.62 49.91
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU 7F32 AMD EPYC 7F32 EPYC 7F32 2 4 6 8 10 6.54145 6.54697 6.50887 MIN: 6.49 MIN: 6.48 MIN: 6.42 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 2 7F32 AMD EPYC 7F32 EPYC 7F32 20 40 60 80 100 80.50 80.89 80.97 1. (CXX) g++ options: -O3 -fPIC -lm
FFmpeg Encoder: libx264 - Scenario: Upload OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Upload 7F32 AMD EPYC 7F32 EPYC 7F32 50 100 150 200 250 228.97 228.16 227.67 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 142.5 143.3 143.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
Stress-NG Test: Malloc OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Malloc 7F32 AMD EPYC 7F32 2M 4M 6M 8M 10M 8973004.75 8923251.96 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
OpenVINO Model: Vehicle Detection FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16-INT8 - Device: CPU 7F32 AMD EPYC 7F32 50 100 150 200 250 220.62 219.40 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
OpenVINO Model: Vehicle Detection FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16-INT8 - Device: CPU 7F32 AMD EPYC 7F32 4 8 12 16 20 18.12 18.22 MIN: 15.25 / MAX: 30.49 MIN: 14.76 / MAX: 32.34 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
FFmpeg Encoder: libx264 - Scenario: Upload OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Upload 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 11.03 11.07 11.09 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Stress-NG Test: System V Message Passing OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: System V Message Passing 7F32 AMD EPYC 7F32 700K 1400K 2100K 2800K 3500K 3167586.04 3150570.51 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
miniBUDE Implementation: OpenMP - Input Deck: BM1 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 7F32 AMD EPYC 7F32 EPYC 7F32 60 120 180 240 300 269.32 269.52 270.67 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.1 Model: inception-v3 7F32 AMD EPYC 7F32 9 18 27 36 45 38.88 38.69 MIN: 38.02 / MAX: 64.74 MIN: 37.64 / MAX: 67.88 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
miniBUDE Implementation: OpenMP - Input Deck: BM1 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 10.77 10.78 10.83 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenVINO Model: Face Detection FP16 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Face Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 0.459 0.918 1.377 1.836 2.295 2.04 2.03 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
GraphicsMagick Operation: Enhanced OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Enhanced 7F32 AMD EPYC 7F32 EPYC 7F32 50 100 150 200 250 206 206 205 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenVINO Model: Face Detection FP16-INT8 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Face Detection FP16-INT8 - Device: CPU 7F32 AMD EPYC 7F32 300 600 900 1200 1500 1461.33 1468.22 MIN: 1387.48 / MAX: 1520.08 MIN: 1444.65 / MAX: 1522.72 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: BMW27 - Compute: CPU-Only 7F32 AMD EPYC 7F32 30 60 90 120 150 148.42 149.11
SMHasher Hash: FarmHash32 x86_64 AVX OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX 7F32 AMD EPYC 7F32 EPYC 7F32 5K 10K 15K 20K 25K 24497.14 24384.85 24494.95 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
nginx Connections: 200 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 200 7F32 AMD EPYC 7F32 11K 22K 33K 44K 55K 52189.35 51950.16 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenRadioss Model: Bumper Beam OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Bumper Beam 7F32 AMD EPYC 7F32 EPYC 7F32 40 80 120 160 200 173.25 174.04 173.95
FFmpeg Encoder: libx265 - Scenario: Upload OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Upload 7F32 AMD EPYC 7F32 EPYC 7F32 40 80 120 160 200 176.34 175.58 176.33 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
srsRAN Test: OFDM_Test OpenBenchmarking.org Samples / Second, More Is Better srsRAN 22.04.1 Test: OFDM_Test 7F32 AMD EPYC 7F32 EPYC 7F32 20M 40M 60M 80M 100M 115900000 116100000 115600000 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 40 80 120 160 200 159.06 158.39
FFmpeg Encoder: libx265 - Scenario: Upload OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Upload 7F32 AMD EPYC 7F32 EPYC 7F32 4 8 12 16 20 14.32 14.38 14.32 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 6 12 18 24 30 25.12 25.22
Cpuminer-Opt Algorithm: Skeincoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Skeincoin 7F32 AMD EPYC 7F32 EPYC 7F32 15K 30K 45K 60K 75K 68400 68130 68290 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 80 160 240 320 400 354.9 355.1 353.7 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 7F32 AMD EPYC 7F32 EPYC 7F32 60 120 180 240 300 270.91 270.07 271.09 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 10.84 10.80 10.84 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
7-Zip Compression Test: Decompression Rating OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating 7F32 AMD EPYC 7F32 EPYC 7F32 13K 26K 39K 52K 65K 58881 58704 58662 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
GraphicsMagick Operation: Noise-Gaussian OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Noise-Gaussian 7F32 AMD EPYC 7F32 EPYC 7F32 60 120 180 240 300 272 271 271 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
FFmpeg Encoder: libx265 - Scenario: Video On Demand OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Video On Demand 7F32 AMD EPYC 7F32 EPYC 7F32 7 14 21 28 35 29.86 29.97 29.88 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 70 140 210 280 350 328.8 329.5 328.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
FFmpeg Encoder: libx265 - Scenario: Video On Demand OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Video On Demand 7F32 AMD EPYC 7F32 EPYC 7F32 60 120 180 240 300 253.64 252.74 253.56 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Stress-NG Test: SENDFILE OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: SENDFILE 7F32 AMD EPYC 7F32 30K 60K 90K 120K 150K 149917.15 150426.21 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU 7F32 AMD EPYC 7F32 EPYC 7F32 0.7858 1.5716 2.3574 3.1432 3.929 3.48771 3.49229 3.48079 MIN: 3.4 MIN: 3.36 MIN: 3.38 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
srsRAN Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 20 40 60 80 100 95.4 95.3 95.1 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.4 WAV To FLAC 7F32 AMD EPYC 7F32 EPYC 7F32 5 10 15 20 25 21.32 21.28 21.34 1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm
WebP Image Encode Encode Settings: Quality 100, Highest Compression OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Highest Compression 7F32 AMD EPYC 7F32 EPYC 7F32 0.7178 1.4356 2.1534 2.8712 3.589 3.19 3.18 3.18 1. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread
SMHasher Hash: MeowHash x86_64 AES-NI OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI 7F32 AMD EPYC 7F32 EPYC 7F32 7K 14K 21K 28K 35K 34310.93 34206.07 34313.46 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 70 140 210 280 350 327.7 327.5 328.5 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
WebP Image Encode Encode Settings: Quality 100 OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100 7F32 AMD EPYC 7F32 EPYC 7F32 3 6 9 12 15 9.92 9.93 9.90 1. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread
Cpuminer-Opt Algorithm: scrypt OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: scrypt 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 145.69 146.09 145.88 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
FFmpeg Encoder: libx264 - Scenario: Platform OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Platform 7F32 AMD EPYC 7F32 EPYC 7F32 10 20 30 40 50 42.03 42.14 42.05 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
BRL-CAD VGR Performance Metric OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.32.6 VGR Performance Metric 7F32 AMD EPYC 7F32 30K 60K 90K 120K 150K 125073 124747 1. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -pthread -ldl -lm
FFmpeg Encoder: libx264 - Scenario: Platform OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Platform 7F32 AMD EPYC 7F32 EPYC 7F32 40 80 120 160 200 180.21 179.76 180.15 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Cpuminer-Opt Algorithm: Triple SHA-256, Onecoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Triple SHA-256, Onecoin 7F32 AMD EPYC 7F32 EPYC 7F32 30K 60K 90K 120K 150K 132610 132340 132280 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Timed PHP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 8.1.9 Time To Compile 7F32 AMD EPYC 7F32 EPYC 7F32 15 30 45 60 75 65.10 65.20 65.04
GraphicsMagick Operation: Swirl OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Swirl 7F32 AMD EPYC 7F32 EPYC 7F32 100 200 300 400 500 444 443 444 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Stress-NG Test: Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Vector Math 7F32 AMD EPYC 7F32 8K 16K 24K 32K 40K 39495.75 39578.06 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
OpenVINO Model: Face Detection FP16 - Device: CPU OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Face Detection FP16 - Device: CPU 7F32 AMD EPYC 7F32 400 800 1200 1600 2000 1953.96 1957.78 MIN: 1891.5 / MAX: 2037.23 MIN: 1879.68 / MAX: 2023.26 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
WebP Image Encode Encode Settings: Default OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Default 7F32 AMD EPYC 7F32 EPYC 7F32 4 8 12 16 20 15.81 15.83 15.80 1. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread
srsRAN Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 12 24 36 48 60 53.6 53.6 53.5 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
Scikit-Learn Benchmark: MNIST Dataset OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.1.3 Benchmark: MNIST Dataset 7F32 AMD EPYC 7F32 30 60 90 120 150 119.78 119.57
OpenVINO Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU 7F32 AMD EPYC 7F32 1400 2800 4200 5600 7000 6443.59 6432.39 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU 7F32 EPYC 7F32 0.8867 1.7734 2.6601 3.5468 4.4335 3.94084 3.93432 MIN: 3.85 MIN: 3.83 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Facebook RocksDB Test: Update Random OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 7.5.3 Test: Update Random 7F32 AMD EPYC 7F32 80K 160K 240K 320K 400K 352648 352075 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 12 24 36 48 60 54.95 54.87
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream 7F32 AMD EPYC 7F32 4 8 12 16 20 18.20 18.22
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Semaphores 7F32 AMD EPYC 7F32 300K 600K 900K 1200K 1500K 1192572.08 1194194.95 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 5 10 15 20 25 22.57 22.60
EnCodec Target Bandwidth: 24 kbps OpenBenchmarking.org Seconds, Fewer Is Better EnCodec 0.1.1 Target Bandwidth: 24 kbps 7F32 AMD EPYC 7F32 9 18 27 36 45 39.82 39.87
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU 7F32 EPYC 7F32 1.0941 2.1882 3.2823 4.3764 5.4705 4.86267 4.85806 MIN: 4.83 MIN: 4.82 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Cpuminer-Opt Algorithm: Deepcoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Deepcoin 7F32 AMD EPYC 7F32 EPYC 7F32 1600 3200 4800 6400 8000 7488.34 7487.49 7494.50 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Facebook RocksDB Test: Read Random Write Random OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 7.5.3 Test: Read Random Write Random 7F32 AMD EPYC 7F32 300K 600K 900K 1200K 1500K 1195887 1196904 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM 7F32 AMD EPYC 7F32 EPYC 7F32 80 160 240 320 400 354.0 354.0 353.7 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm
SMHasher Hash: t1ha2_atonce OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: t1ha2_atonce 7F32 AMD EPYC 7F32 EPYC 7F32 3K 6K 9K 12K 15K 14594.36 14583.40 14595.36 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
Cpuminer-Opt Algorithm: Myriad-Groestl OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Myriad-Groestl 7F32 AMD EPYC 7F32 EPYC 7F32 3K 6K 9K 12K 15K 12630 12640 12640 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
FFmpeg Encoder: libx264 - Scenario: Video On Demand OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Video On Demand 7F32 AMD EPYC 7F32 EPYC 7F32 10 20 30 40 50 41.94 41.96 41.93 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Stress-NG Test: Atomic OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Atomic 7F32 AMD EPYC 7F32 80K 160K 240K 320K 400K 392155.78 392425.59 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
Stress-NG Test: NUMA OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: NUMA 7F32 AMD EPYC 7F32 50 100 150 200 250 236.87 236.71 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lrt -lsctp -lz -pthread
FFmpeg Encoder: libx264 - Scenario: Video On Demand OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Video On Demand 7F32 AMD EPYC 7F32 EPYC 7F32 40 80 120 160 200 180.62 180.54 180.64 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
SMHasher Hash: SHA3-256 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: SHA3-256 7F32 AMD EPYC 7F32 EPYC 7F32 30 60 90 120 150 136.24 136.22 136.17 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
Cpuminer-Opt Algorithm: Quad SHA-256, Pyrite OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Quad SHA-256, Pyrite 7F32 AMD EPYC 7F32 EPYC 7F32 14K 28K 42K 56K 70K 66260 66230 66230 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
SMHasher Hash: Spooky32 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: Spooky32 7F32 AMD EPYC 7F32 EPYC 7F32 3K 6K 9K 12K 15K 13220.33 13216.41 13214.70 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: FarmHash128 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: FarmHash128 7F32 AMD EPYC 7F32 EPYC 7F32 3K 6K 9K 12K 15K 14307.94 14312.80 14311.97 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: wyhash OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: wyhash 7F32 AMD EPYC 7F32 EPYC 7F32 5K 10K 15K 20K 25K 21041.15 21040.94 21044.04 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: fasthash32 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: fasthash32 7F32 AMD EPYC 7F32 EPYC 7F32 1300 2600 3900 5200 6500 6027.12 6027.73 6027.75 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream 7F32 AMD EPYC 7F32 40 80 120 160 200 176.85 176.84
Mobile Neural Network Model: nasnet OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.1 Model: nasnet 7F32 AMD EPYC 7F32 6 12 18 24 30 23.65 23.65 MIN: 23.33 / MAX: 44.97 MIN: 21.78 / MAX: 47.46 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p 7F32 AMD EPYC 7F32 EPYC 7F32 0.1215 0.243 0.3645 0.486 0.6075 0.54 0.54 0.54 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K 7F32 AMD EPYC 7F32 EPYC 7F32 0.0428 0.0856 0.1284 0.1712 0.214 0.19 0.19 0.19 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
WebP2 Image Encode Encode Settings: Quality 100, Lossless Compression OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 100, Lossless Compression 7F32 AMD EPYC 7F32 EPYC 7F32 0.0023 0.0046 0.0069 0.0092 0.0115 0.01 0.01 0.01 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl -lpthread
WebP2 Image Encode Encode Settings: Quality 100, Compression Effort 5 OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 100, Compression Effort 5 7F32 AMD EPYC 7F32 EPYC 7F32 0.7313 1.4626 2.1939 2.9252 3.6565 3.25 3.25 3.25 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl -lpthread
WebP2 Image Encode Encode Settings: Quality 95, Compression Effort 7 OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 95, Compression Effort 7 7F32 AMD EPYC 7F32 EPYC 7F32 0.0113 0.0226 0.0339 0.0452 0.0565 0.05 0.05 0.05 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl -lpthread
WebP2 Image Encode Encode Settings: Quality 75, Compression Effort 7 OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 75, Compression Effort 7 7F32 AMD EPYC 7F32 EPYC 7F32 0.027 0.054 0.081 0.108 0.135 0.12 0.12 0.12 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl -lpthread
WebP Image Encode Encode Settings: Quality 100, Lossless, Highest Compression OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless, Highest Compression 7F32 AMD EPYC 7F32 EPYC 7F32 0.1305 0.261 0.3915 0.522 0.6525 0.58 0.58 0.58 1. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread
JPEG XL libjxl Input: JPEG - Quality: 100 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: JPEG - Quality: 100 7F32 AMD EPYC 7F32 EPYC 7F32 0.1305 0.261 0.3915 0.522 0.6525 0.58 0.58 0.58 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic
SMHasher Hash: MeowHash x86_64 AES-NI OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI 7F32 AMD EPYC 7F32 EPYC 7F32 14 28 42 56 70 63.11 63.07 63.12 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: t1ha0_aes_avx2 x86_64 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 7F32 AMD EPYC 7F32 EPYC 7F32 9 18 27 36 45 37.77 37.77 37.77 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: FarmHash32 x86_64 AVX OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX 7F32 AMD EPYC 7F32 EPYC 7F32 11 22 33 44 55 47.68 47.68 47.68 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: t1ha2_atonce OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: t1ha2_atonce 7F32 AMD EPYC 7F32 EPYC 7F32 9 18 27 36 45 38.50 38.50 38.50 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: FarmHash128 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: FarmHash128 7F32 AMD EPYC 7F32 EPYC 7F32 16 32 48 64 80 70.39 70.29 70.29 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: fasthash32 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: fasthash32 7F32 AMD EPYC 7F32 EPYC 7F32 9 18 27 36 45 40.69 40.68 40.68 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: Spooky32 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: Spooky32 7F32 AMD EPYC 7F32 EPYC 7F32 13 26 39 52 65 56.14 55.92 56.22 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: SHA3-256 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: SHA3-256 7F32 AMD EPYC 7F32 EPYC 7F32 600 1200 1800 2400 3000 2840.32 2848.54 2843.50 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
SMHasher Hash: wyhash OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: wyhash 7F32 AMD EPYC 7F32 EPYC 7F32 7 14 21 28 35 28.39 28.39 28.39 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
Phoronix Test Suite v10.8.5