2 x Intel Xeon Gold 5220R testing with a TYAN S7106 (V2.01.B40 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2304025-NE-EEE49872408 eee - Phoronix Test Suite eee 2 x Intel Xeon Gold 5220R testing with a TYAN S7106 (V2.01.B40 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2304025-NE-EEE49872408&sro&grs .
eee Processor Motherboard Chipset Memory Disk Graphics Monitor Network OS Kernel Desktop Display Server Compiler File-System Screen Resolution a b bb 2 x Intel Xeon Gold 5220R @ 3.90GHz (36 Cores / 72 Threads) TYAN S7106 (V2.01.B40 BIOS) Intel Sky Lake-E DMI3 Registers 94GB 500GB Samsung SSD 860 ASPEED VE228 2 x Intel I210 + 2 x QLogic cLOM8214 1/10GbE Ubuntu 20.04 6.1.0-phx (x86_64) GNOME Shell 3.36.9 X Server 1.20.13 GCC 9.4.0 ext4 1920x1080 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x5003302 Python Details - Python 2.7.18 + Python 3.8.10 Security Details - itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Stuffing + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled
eee tensorflow: CPU - 32 - GoogLeNet onednn: Convolution Batch Shapes Auto - f32 - CPU pgbench: 100 - 50 - Read Write - Average Latency pgbench: 100 - 50 - Read Write onednn: IP Shapes 3D - bf16bf16bf16 - CPU memcached: 1:100 memcached: 1:5 memcached: 1:10 compress-zstd: 3, Long Mode - Compression Speed onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU onednn: IP Shapes 3D - f32 - CPU ffmpeg: libx265 - Live ffmpeg: libx265 - Live rocksdb: Rand Fill Sync john-the-ripper: HMAC-SHA512 svt-av1: Preset 12 - Bosphorus 4K onednn: IP Shapes 1D - u8s8f32 - CPU compress-zstd: 8 - Compression Speed clickhouse: 100M Rows Hits Dataset, First Run / Cold Cache clickhouse: 100M Rows Hits Dataset, Second Run compress-zstd: 19, Long Mode - Compression Speed compress-zstd: 12 - Compression Speed compress-zstd: 3 - Compression Speed build-llvm: Unix Makefiles rocksdb: Seq Fill svt-av1: Preset 4 - Bosphorus 4K compress-zstd: 8, Long Mode - Compression Speed rocksdb: Rand Fill deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream svt-av1: Preset 12 - Bosphorus 1080p tensorflow: CPU - 16 - AlexNet clickhouse: 100M Rows Hits Dataset, Third Run vvenc: Bosphorus 4K - Fast deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Stream deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Stream specfem3d: Mount St. Helens rocksdb: Update Rand svt-av1: Preset 4 - Bosphorus 1080p specfem3d: Water-layered Halfspace deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream build2: Time To Compile deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream onednn: IP Shapes 3D - u8s8f32 - CPU svt-av1: Preset 8 - Bosphorus 1080p pgbench: 1 - 50 - Read Only nginx: 500 uvg266: Bosphorus 4K - Ultra Fast nginx: 200 vvenc: Bosphorus 1080p - Fast svt-av1: Preset 8 - Bosphorus 4K uvg266: Bosphorus 1080p - Very Fast specfem3d: Tomographic Model onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU vvenc: Bosphorus 1080p - Faster uvg266: Bosphorus 1080p - Super Fast deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream uvg266: Bosphorus 4K - Super Fast tensorflow: CPU - 256 - GoogLeNet build-llvm: Ninja specfem3d: Layered Halfspace ffmpeg: libx264 - Live ffmpeg: libx264 - Live ffmpeg: libx265 - Upload openssl: SHA256 ffmpeg: libx265 - Upload deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream onednn: Deconvolution Batch shapes_1d - f32 - CPU tensorflow: CPU - 64 - AlexNet openssl: RSA4096 rocksdb: Read Rand Write Rand specfem3d: Homogeneous Halfspace compress-zstd: 19 - Decompression Speed compress-zstd: 19, Long Mode - Decompression Speed build-godot: Time To Compile vvenc: Bosphorus 4K - Faster uvg266: Bosphorus 1080p - Slow onednn: IP Shapes 1D - f32 - CPU build-ffmpeg: Time To Compile onednn: Recurrent Neural Network Training - f32 - CPU deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream compress-zstd: 12 - Decompression Speed deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream uvg266: Bosphorus 1080p - Medium embree: Pathtracer - Crown tensorflow: CPU - 16 - GoogLeNet deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU svt-av1: Preset 13 - Bosphorus 4K compress-zstd: 19 - Compression Speed john-the-ripper: Blowfish rocksdb: Read While Writing compress-zstd: 8, Long Mode - Decompression Speed embree: Pathtracer ISPC - Crown compress-zstd: 3 - Decompression Speed tensorflow: CPU - 256 - ResNet-50 tensorflow: CPU - 64 - ResNet-50 compress-zstd: 3, Long Mode - Decompression Speed ffmpeg: libx264 - Video On Demand john-the-ripper: WPA PSK ffmpeg: libx264 - Video On Demand deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream tensorflow: CPU - 32 - AlexNet john-the-ripper: MD5 onednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPU onednn: Recurrent Neural Network Training - u8s8f32 - CPU nginx: 1000 blender: BMW27 - CPU-Only embree: Pathtracer - Asian Dragon Obj onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU uvg266: Bosphorus 4K - Slow tensorflow: CPU - 512 - ResNet-50 ffmpeg: libx264 - Upload uvg266: Bosphorus 1080p - Ultra Fast openssl: SHA512 svt-av1: Preset 13 - Bosphorus 1080p ffmpeg: libx264 - Upload embree: Pathtracer ISPC - Asian Dragon tensorflow: CPU - 512 - GoogLeNet onednn: Recurrent Neural Network Inference - u8s8f32 - CPU embree: Pathtracer - Asian Dragon pgbench: 1 - 50 - Read Write - Average Latency openssl: AES-128-GCM compress-zstd: 8 - Decompression Speed uvg266: Bosphorus 4K - Very Fast tensorflow: CPU - 256 - AlexNet onednn: Recurrent Neural Network Inference - f32 - CPU tensorflow: CPU - 512 - AlexNet pgbench: 1 - 50 - Read Write deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream ffmpeg: libx264 - Platform deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream ffmpeg: libx265 - Video On Demand openssl: RSA4096 ffmpeg: libx264 - Platform embree: Pathtracer ISPC - Asian Dragon Obj ffmpeg: libx265 - Platform pgbench: 100 - 50 - Read Only uvg266: Bosphorus 4K - Medium ffmpeg: libx265 - Video On Demand deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream rocksdb: Rand Read tensorflow: CPU - 32 - ResNet-50 draco: Church Facade ffmpeg: libx265 - Platform nginx: 100 openssl: AES-256-GCM draco: Lion deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Stream deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Stream deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream blender: Pabellon Barcelona - CPU-Only blender: Barbershop - CPU-Only onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU blender: Fishy Cat - CPU-Only deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream onednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPU tensorflow: CPU - 64 - GoogLeNet onednn: Deconvolution Batch shapes_3d - f32 - CPU john-the-ripper: bcrypt blender: Classroom - CPU-Only build-nodejs: Time To Compile tensorflow: CPU - 16 - ResNet-50 onednn: IP Shapes 1D - bf16bf16bf16 - CPU onednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPU deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream openssl: ChaCha20 deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Stream openssl: ChaCha20-Poly1305 deepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Stream pgbench: 100 - 50 - Read Only - Average Latency pgbench: 1 - 50 - Read Only - Average Latency apache: 100 a b bb 43.32 3.5297 12.67 3946 4.0573 1290869.25 2007905.01 1719545.48 325.9 4.39463 3.45599 125.374756424 40.28 6971 91245000 110.282 1.63878 579.4 163.13 205.05 6.71 146.2 1664.3 439.12 259650 2.55 282.7 251487 8.8739 112.5065 232.248 83.2 206.05 3.657 403.0469 44.6206 24.148005301 233375 6.223 60.237167581 18.3814 977.294 100.27 18.4172 1.29978 85.659 1068707 141761.59 21.8 147630.66 9.605 36.413 42.93 25.349439354 786.919 16.729 44.64 40.3347 446.1672 20.29 103.29 343.242 63.682105721 33.20 152.11 237.23567868 9173311330 10.64 65.3784 15.2836 14.7513 146.93 8029.4 2228918 31.379210956 840.6 826 232.641 5.991 18.3 1.73354 33.48 1384.14 23.4331 987.6 42.6499 20.04 28.6383 50.71 51.1117 19.5562 0.693934 110.638 12.8 48756 4910200 1105.7 29.7114 1102.7 29.63 22.2 1137 38.03 212818 199.18 34.7145 28.7927 109.53 4691000 9.56667 1379.97 139200.66 56.07 29.595 1374.37 7.91 31.1 10.44 48.17 10145725330 200.802 241.764027512 37.8838 109.76 788.804 33.0349 61.842 159151287770 1093.3 18.07 212.65 786.003 227.16 809 100.5244 38.19 178.9198 366.57 534883.8 198.347553986 32.4996 20.45 994655 8.83 20.66 965.2134 115209715 18.25 9410 370.338638914 152180.63 118250262800 6694 8.668 115.3529 8.8732 112.5355 99.4868 180.7166 191.67 579.52 0.563515 74.47 115.1583 8.6827 8.67071 77.76 2.74347 48821 159.56 309.57 15.11 5.69353 6.35319 196.5611 150176760350 351.8332 91.4064 51.1119 139.5608 128.7775 54.3343 80516466550 18.3878 0.05 0.047 64.04 4.78799 13.85 3610 4.30617 1550585.25 1672007.69 1478446.78 283 4.90144 3.09364 115.765549178 43.62 91970000 117.742 1.74816 545.9 157.77 211.36 6.45 153.9 1640.6 439.282 2.608 289.9 235.271 84.48 210.46 3.579 24.125490298 6.344 61.367356552 99.151 1.28042 86.951 1052848 21.48 9.641 36.225 42.62 25.268860658 797.591 16.897 45.1 20.33 104.62 339.912 64.462549436 32.80 153.96 238.65 9102032020 10.58 14.8033 146.48 8026.7 31.089935816 846.7 833.2 234.221 6.026 18.45 1.7178 33.273 1386.62 996.4 20.05 28.4938 50.96 0.688332 111.154 12.8 48405 1097.9 29.5023 1095.1 22.27 1144.5 38.16 213811 198.50 109.74 4719000 9.51019 1385.41 29.4408 1381.33 7.95 10.42 48.24 10167598450 201.732 242.24275493 37.7104 785.39 33.1778 61.656 159781972650 1093.7 18.11 212.5 788.652 227.85 811 38.07 367.613657159 536533.8 198.960340617 32.5565 20.52 991262 8.82 20.61 18.31 369.219786333 118596420950 0.562744 8.68912 77.64 2.73821 48864 309.311 15.1 5.68728 6.35868 150074005090 80512557330 0.05 0.047 63.78 4.77498 16.497 3031 3.33073 1587757.54 1937351.98 1700246.18 310.5 4.35862 3.17173 124.51 40.56 7527 98271000 117.555 1.69336 572.3 154.30 200.24 6.79 151.4 1584.1 419.847 249987 2.647 279.4 243680 9.1226 109.4518 228.891 82.22 211.12 3.634 394.4534 45.5898 24.639822177 228703 6.222 60.513328909 18.7109 961.1346 100.747 18.7079 1.28253 86.484 1067954 143874.15 21.62 149820.05 9.746 35.893 42.32 25.0077492 789.808 16.675 45.23 40.8629 440.4085 20.07 104.2 338.885 64.480914525 32.83 153.822107334 235.941449418 9206384930 10.70 66.0925 15.1185 14.6464 145.41 8109.8 2251517 31.065733738 849 834.2 232.02 5.97 18.47 1.73257 33.178 1374.19 23.6443 996.5 42.2697 20.22 28.7455 51.15 50.6953 19.7168 0.691037 110.286 12.9 48562 4875480 1100.7 29.5362 1095.8 29.43 22.35 1140.2 38.28 214200 197.896164664 34.9308 28.6153 109.07 4712000 9.51252 1377.45 139997.58 55.76 29.4853 1380.03 7.91 30.95 10.47 48.4 10193373610 201.444 241.13 37.7176 109.27 787.693 33.1711 61.913 159160934620 1097.6 18.14 213.32 789.003 227 808 100.1536 38.21 179.5686 366.29 534630.5 198.27 32.4438 20.50 993660 8.85 20.68 961.9707 114830146 18.3 9381 369.540182188 151733.73 118499993460 6713 8.6925 115.0291 8.8977 112.2279 99.7544 180.2832 192.11 580.85 0.562273 74.63 115.4054 8.6642 8.67211 77.79 2.7383 48912 159.27 309.086 15.09 5.68651 6.35249 196.7412 150059242840 352.0435 91.353 51.0891 139.499 128.8102 54.3298 80513755000 18.3879 0.05 0.047 OpenBenchmarking.org
TensorFlow Device: CPU - Batch Size: 32 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: GoogLeNet a b bb 14 28 42 56 70 43.32 64.04 63.78
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU a b bb 1.0773 2.1546 3.2319 4.3092 5.3865 3.52970 4.78799 4.77498 MIN: 2.77 MIN: 2.75 MIN: 2.76 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
PostgreSQL Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average Latency a b bb 4 8 12 16 20 12.67 13.85 16.50 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL Scaling Factor: 100 - Clients: 50 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 50 - Mode: Read Write a b bb 800 1600 2400 3200 4000 3946 3610 3031 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
oneDNN Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU a b bb 0.9689 1.9378 2.9067 3.8756 4.8445 4.05730 4.30617 3.33073 MIN: 2.84 MIN: 2.98 MIN: 2.9 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Memcached Set To Get Ratio: 1:100 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:100 a b bb 300K 600K 900K 1200K 1500K 1290869.25 1550585.25 1587757.54 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Memcached Set To Get Ratio: 1:5 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:5 a b bb 400K 800K 1200K 1600K 2000K 2007905.01 1672007.69 1937351.98 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Memcached Set To Get Ratio: 1:10 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:10 a b bb 400K 800K 1200K 1600K 2000K 1719545.48 1478446.78 1700246.18 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Zstd Compression Compression Level: 3, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 3, Long Mode - Compression Speed a b bb 70 140 210 280 350 325.9 283.0 310.5 1. (CC) gcc options: -O3 -pthread -lz -llzma
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU a b bb 1.1028 2.2056 3.3084 4.4112 5.514 4.39463 4.90144 4.35862 MIN: 2.48 MIN: 2.4 MIN: 2.47 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 3.1 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU a b bb 0.7776 1.5552 2.3328 3.1104 3.888 3.45599 3.09364 3.17173 MIN: 1.83 MIN: 1.81 MIN: 1.84 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
FFmpeg Encoder: libx265 - Scenario: Live OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Live a b bb 30 60 90 120 150 125.37 115.77 124.51 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx265 - Scenario: Live OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Live a b bb 10 20 30 40 50 40.28 43.62 40.56 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
RocksDB Test: Random Fill Sync OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Random Fill Sync a bb 1600 3200 4800 6400 8000 6971 7527 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
John The Ripper Test: HMAC-SHA512 OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: HMAC-SHA512 a b bb 20M 40M 60M 80M 100M 91245000 91970000 98271000 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lrt -lz -ldl -lcrypt -lbz2
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 12 - Input: Bosphorus 4K a b bb 30 60 90 120 150 110.28 117.74 117.56 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
oneDNN Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU a b bb 0.3933 0.7866 1.1799 1.5732 1.9665 1.63878 1.74816 1.69336 MIN: 1.5 MIN: 1.56 MIN: 1.56 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Zstd Compression Compression Level: 8 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8 - Compression Speed a b bb 130 260 390 520 650 579.4 545.9 572.3 1. (CC) gcc options: -O3 -pthread -lz -llzma
ClickHouse 100M Rows Hits Dataset, First Run / Cold Cache OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache a b bb 40 80 120 160 200 163.13 157.77 154.30 MIN: 16.41 / MAX: 1200 MIN: 16.57 / MAX: 1621.62 MIN: 15.97 / MAX: 1578.95
ClickHouse 100M Rows Hits Dataset, Second Run OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run a b bb 50 100 150 200 250 205.05 211.36 200.24 MIN: 18.52 / MAX: 1363.64 MIN: 18.65 / MAX: 2500 MIN: 18.21 / MAX: 1714.29
Zstd Compression Compression Level: 19, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Compression Speed a b bb 2 4 6 8 10 6.71 6.45 6.79 1. (CC) gcc options: -O3 -pthread -lz -llzma
Zstd Compression Compression Level: 12 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 12 - Compression Speed a b bb 30 60 90 120 150 146.2 153.9 151.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
Zstd Compression Compression Level: 3 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 3 - Compression Speed a b bb 400 800 1200 1600 2000 1664.3 1640.6 1584.1 1. (CC) gcc options: -O3 -pthread -lz -llzma
Timed LLVM Compilation Build System: Unix Makefiles OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Unix Makefiles a b bb 100 200 300 400 500 439.12 439.28 419.85
RocksDB Test: Sequential Fill OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Sequential Fill a bb 60K 120K 180K 240K 300K 259650 249987 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 4 - Input: Bosphorus 4K a b bb 0.5956 1.1912 1.7868 2.3824 2.978 2.550 2.608 2.647 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Zstd Compression Compression Level: 8, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8, Long Mode - Compression Speed a b bb 60 120 180 240 300 282.7 289.9 279.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
RocksDB Test: Random Fill OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Random Fill a bb 50K 100K 150K 200K 250K 251487 243680 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream a bb 3 6 9 12 15 8.8739 9.1226
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream a bb 30 60 90 120 150 112.51 109.45
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 12 - Input: Bosphorus 1080p a b bb 50 100 150 200 250 232.25 235.27 228.89 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
TensorFlow Device: CPU - Batch Size: 16 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: AlexNet a b bb 20 40 60 80 100 83.20 84.48 82.22
ClickHouse 100M Rows Hits Dataset, Third Run OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run a b bb 50 100 150 200 250 206.05 210.46 211.12 MIN: 18.75 / MAX: 1090.91 MIN: 18.39 / MAX: 1276.6 MIN: 18.35 / MAX: 1875
VVenC Video Input: Bosphorus 4K - Video Preset: Fast OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.7 Video Input: Bosphorus 4K - Video Preset: Fast a b bb 0.8228 1.6456 2.4684 3.2912 4.114 3.657 3.579 3.634 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto -lpthread
Neural Magic DeepSparse Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream a bb 90 180 270 360 450 403.05 394.45
Neural Magic DeepSparse Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream a bb 10 20 30 40 50 44.62 45.59
SPECFEM3D Model: Mount St. Helens OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Mount St. Helens a b bb 6 12 18 24 30 24.15 24.13 24.64 1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
RocksDB Test: Update Random OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Update Random a bb 50K 100K 150K 200K 250K 233375 228703 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 4 - Input: Bosphorus 1080p a b bb 2 4 6 8 10 6.223 6.344 6.222 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
SPECFEM3D Model: Water-layered Halfspace OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Water-layered Halfspace a b bb 14 28 42 56 70 60.24 61.37 60.51 1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
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 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a bb 5 10 15 20 25 18.38 18.71
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 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a bb 200 400 600 800 1000 977.29 961.13
Build2 Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.15 Time To Compile a b bb 20 40 60 80 100 100.27 99.15 100.75
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 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream a bb 5 10 15 20 25 18.42 18.71
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU a b bb 0.2925 0.585 0.8775 1.17 1.4625 1.29978 1.28042 1.28253 MIN: 1.08 MIN: 1.07 MIN: 1.1 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 8 - Input: Bosphorus 1080p a b bb 20 40 60 80 100 85.66 86.95 86.48 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
PostgreSQL Scaling Factor: 1 - Clients: 50 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 1 - Clients: 50 - Mode: Read Only a b bb 200K 400K 600K 800K 1000K 1068707 1052848 1067954 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
nginx Connections: 500 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 a bb 30K 60K 90K 120K 150K 141761.59 143874.15 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
uvg266 Video Input: Bosphorus 4K - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast a b bb 5 10 15 20 25 21.80 21.48 21.62
nginx Connections: 200 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 200 a bb 30K 60K 90K 120K 150K 147630.66 149820.05 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
VVenC Video Input: Bosphorus 1080p - Video Preset: Fast OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.7 Video Input: Bosphorus 1080p - Video Preset: Fast a b bb 3 6 9 12 15 9.605 9.641 9.746 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto -lpthread
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 8 - Input: Bosphorus 4K a b bb 8 16 24 32 40 36.41 36.23 35.89 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
uvg266 Video Input: Bosphorus 1080p - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast a b bb 10 20 30 40 50 42.93 42.62 42.32
SPECFEM3D Model: Tomographic Model OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Tomographic Model a b bb 6 12 18 24 30 25.35 25.27 25.01 1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU a b bb 200 400 600 800 1000 786.92 797.59 789.81 MIN: 779.8 MIN: 783.04 MIN: 783.36 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
VVenC Video Input: Bosphorus 1080p - Video Preset: Faster OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.7 Video Input: Bosphorus 1080p - Video Preset: Faster a b bb 4 8 12 16 20 16.73 16.90 16.68 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto -lpthread
uvg266 Video Input: Bosphorus 1080p - Video Preset: Super Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Super Fast a b bb 10 20 30 40 50 44.64 45.10 45.23
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream a bb 9 18 27 36 45 40.33 40.86
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream a bb 100 200 300 400 500 446.17 440.41
uvg266 Video Input: Bosphorus 4K - Video Preset: Super Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast a b bb 5 10 15 20 25 20.29 20.33 20.07
TensorFlow Device: CPU - Batch Size: 256 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 256 - Model: GoogLeNet a b bb 20 40 60 80 100 103.29 104.62 104.20
Timed LLVM Compilation Build System: Ninja OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Ninja a b bb 70 140 210 280 350 343.24 339.91 338.89
SPECFEM3D Model: Layered Halfspace OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Layered Halfspace a b bb 14 28 42 56 70 63.68 64.46 64.48 1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
FFmpeg Encoder: libx264 - Scenario: Live OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Live a b bb 8 16 24 32 40 33.20 32.80 32.83 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx264 - Scenario: Live OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Live a b bb 30 60 90 120 150 152.11 153.96 153.82 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx265 - Scenario: Upload OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Upload a b bb 50 100 150 200 250 237.24 238.65 235.94 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenSSL Algorithm: SHA256 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA256 a b bb 2000M 4000M 6000M 8000M 10000M 9173311330 9102032020 9206384930 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
FFmpeg Encoder: libx265 - Scenario: Upload OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Upload a b bb 3 6 9 12 15 10.64 10.58 10.70 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream a bb 15 30 45 60 75 65.38 66.09
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream a bb 4 8 12 16 20 15.28 15.12
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU a b bb 4 8 12 16 20 14.75 14.80 14.65 MIN: 12.24 MIN: 12.39 MIN: 12.43 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
TensorFlow Device: CPU - Batch Size: 64 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: AlexNet a b bb 30 60 90 120 150 146.93 146.48 145.41
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 a b bb 2K 4K 6K 8K 10K 8029.4 8026.7 8109.8 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
RocksDB Test: Read Random Write Random OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Read Random Write Random a bb 500K 1000K 1500K 2000K 2500K 2228918 2251517 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
SPECFEM3D Model: Homogeneous Halfspace OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Homogeneous Halfspace a b bb 7 14 21 28 35 31.38 31.09 31.07 1. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Zstd Compression Compression Level: 19 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed a b bb 200 400 600 800 1000 840.6 846.7 849.0 1. (CC) gcc options: -O3 -pthread -lz -llzma
Zstd Compression Compression Level: 19, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed a b bb 200 400 600 800 1000 826.0 833.2 834.2 1. (CC) gcc options: -O3 -pthread -lz -llzma
Timed Godot Game Engine Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 4.0 Time To Compile a b bb 50 100 150 200 250 232.64 234.22 232.02
VVenC Video Input: Bosphorus 4K - Video Preset: Faster OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.7 Video Input: Bosphorus 4K - Video Preset: Faster a b bb 2 4 6 8 10 5.991 6.026 5.970 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto -lpthread
uvg266 Video Input: Bosphorus 1080p - Video Preset: Slow OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Slow a b bb 5 10 15 20 25 18.30 18.45 18.47
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU a b bb 0.39 0.78 1.17 1.56 1.95 1.73354 1.71780 1.73257 MIN: 1.65 MIN: 1.64 MIN: 1.64 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Timed FFmpeg Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed FFmpeg Compilation 6.0 Time To Compile a b bb 8 16 24 32 40 33.48 33.27 33.18
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU a b bb 300 600 900 1200 1500 1384.14 1386.62 1374.19 MIN: 1379.57 MIN: 1378.88 MIN: 1367.79 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
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 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream a bb 6 12 18 24 30 23.43 23.64
Zstd Compression Compression Level: 12 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 12 - Decompression Speed a b bb 200 400 600 800 1000 987.6 996.4 996.5 1. (CC) gcc options: -O3 -pthread -lz -llzma
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 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream a bb 10 20 30 40 50 42.65 42.27
uvg266 Video Input: Bosphorus 1080p - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Medium a b bb 5 10 15 20 25 20.04 20.05 20.22
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer - Model: Crown a b bb 7 14 21 28 35 28.64 28.49 28.75 MIN: 28.27 / MAX: 29.08 MIN: 28.14 / MAX: 29 MIN: 28.39 / MAX: 29.15
TensorFlow Device: CPU - Batch Size: 16 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: GoogLeNet a b bb 12 24 36 48 60 50.71 50.96 51.15
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream a bb 12 24 36 48 60 51.11 50.70
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream a bb 5 10 15 20 25 19.56 19.72
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU a b bb 0.1561 0.3122 0.4683 0.6244 0.7805 0.693934 0.688332 0.691037 MIN: 0.69 MIN: 0.68 MIN: 0.68 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 13 - Input: Bosphorus 4K a b bb 20 40 60 80 100 110.64 111.15 110.29 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Zstd Compression Compression Level: 19 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Compression Speed a b bb 3 6 9 12 15 12.8 12.8 12.9 1. (CC) gcc options: -O3 -pthread -lz -llzma
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: Blowfish a b bb 10K 20K 30K 40K 50K 48756 48405 48562 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lrt -lz -ldl -lcrypt -lbz2
RocksDB Test: Read While Writing OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Read While Writing a bb 1.1M 2.2M 3.3M 4.4M 5.5M 4910200 4875480 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Zstd Compression Compression Level: 8, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8, Long Mode - Decompression Speed a b bb 200 400 600 800 1000 1105.7 1097.9 1100.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer ISPC - Model: Crown a b bb 7 14 21 28 35 29.71 29.50 29.54 MIN: 29.3 / MAX: 30.17 MIN: 29.05 / MAX: 30.07 MIN: 29.06 / MAX: 30.1
Zstd Compression Compression Level: 3 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 3 - Decompression Speed a b bb 200 400 600 800 1000 1102.7 1095.1 1095.8 1. (CC) gcc options: -O3 -pthread -lz -llzma
TensorFlow Device: CPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 256 - Model: ResNet-50 a bb 7 14 21 28 35 29.63 29.43
TensorFlow Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 a b bb 5 10 15 20 25 22.20 22.27 22.35
Zstd Compression Compression Level: 3, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 3, Long Mode - Decompression Speed a b bb 200 400 600 800 1000 1137.0 1144.5 1140.2 1. (CC) gcc options: -O3 -pthread -lz -llzma
FFmpeg Encoder: libx264 - Scenario: Video On Demand OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Video On Demand a b bb 9 18 27 36 45 38.03 38.16 38.28 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
John The Ripper Test: WPA PSK OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: WPA PSK a b bb 50K 100K 150K 200K 250K 212818 213811 214200 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lrt -lz -ldl -lcrypt -lbz2
FFmpeg Encoder: libx264 - Scenario: Video On Demand OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Video On Demand a b bb 40 80 120 160 200 199.18 198.50 197.90 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
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 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream a bb 8 16 24 32 40 34.71 34.93
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 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream a bb 7 14 21 28 35 28.79 28.62
TensorFlow Device: CPU - Batch Size: 32 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: AlexNet a b bb 20 40 60 80 100 109.53 109.74 109.07
John The Ripper Test: MD5 OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: MD5 a b bb 1000K 2000K 3000K 4000K 5000K 4691000 4719000 4712000 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lrt -lz -ldl -lcrypt -lbz2
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU a b bb 3 6 9 12 15 9.56667 9.51019 9.51252 MIN: 9.44 MIN: 9.44 MIN: 9.44 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU a b bb 300 600 900 1200 1500 1379.97 1385.41 1377.45 MIN: 1367.35 MIN: 1377.53 MIN: 1371.04 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
nginx Connections: 1000 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 a bb 30K 60K 90K 120K 150K 139200.66 139997.58 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: BMW27 - Compute: CPU-Only a bb 13 26 39 52 65 56.07 55.76
Embree Binary: Pathtracer - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer - Model: Asian Dragon Obj a b bb 7 14 21 28 35 29.60 29.44 29.49 MIN: 29.37 / MAX: 30.06 MIN: 29.24 / MAX: 29.86 MIN: 29.24 / MAX: 29.77
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU a b bb 300 600 900 1200 1500 1374.37 1381.33 1380.03 MIN: 1363.77 MIN: 1376.64 MIN: 1374.88 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
uvg266 Video Input: Bosphorus 4K - Video Preset: Slow OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Slow a b bb 2 4 6 8 10 7.91 7.95 7.91
TensorFlow Device: CPU - Batch Size: 512 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 512 - Model: ResNet-50 a bb 7 14 21 28 35 31.10 30.95
FFmpeg Encoder: libx264 - Scenario: Upload OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Upload a b bb 3 6 9 12 15 10.44 10.42 10.47 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
uvg266 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast a b bb 11 22 33 44 55 48.17 48.24 48.40
OpenSSL Algorithm: SHA512 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA512 a b bb 2000M 4000M 6000M 8000M 10000M 10145725330 10167598450 10193373610 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.4 Encoder Mode: Preset 13 - Input: Bosphorus 1080p a b bb 40 80 120 160 200 200.80 201.73 201.44 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
FFmpeg Encoder: libx264 - Scenario: Upload OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Upload a b bb 50 100 150 200 250 241.76 242.24 241.13 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer ISPC - Model: Asian Dragon a b bb 9 18 27 36 45 37.88 37.71 37.72 MIN: 37.64 / MAX: 38.3 MIN: 37.4 / MAX: 38.13 MIN: 37.47 / MAX: 38.19
TensorFlow Device: CPU - Batch Size: 512 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 512 - Model: GoogLeNet a bb 20 40 60 80 100 109.76 109.27
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU a b bb 200 400 600 800 1000 788.80 785.39 787.69 MIN: 782.96 MIN: 775.34 MIN: 778.68 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer - Model: Asian Dragon a b bb 8 16 24 32 40 33.03 33.18 33.17 MIN: 32.82 / MAX: 33.48 MIN: 32.97 / MAX: 33.5 MIN: 32.95 / MAX: 33.5
PostgreSQL Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average Latency a b bb 14 28 42 56 70 61.84 61.66 61.91 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenSSL Algorithm: AES-128-GCM OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-128-GCM a b bb 30000M 60000M 90000M 120000M 150000M 159151287770 159781972650 159160934620 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Zstd Compression Compression Level: 8 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8 - Decompression Speed a b bb 200 400 600 800 1000 1093.3 1093.7 1097.6 1. (CC) gcc options: -O3 -pthread -lz -llzma
uvg266 Video Input: Bosphorus 4K - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast a b bb 4 8 12 16 20 18.07 18.11 18.14
TensorFlow Device: CPU - Batch Size: 256 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 256 - Model: AlexNet a b bb 50 100 150 200 250 212.65 212.50 213.32
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU a b bb 200 400 600 800 1000 786.00 788.65 789.00 MIN: 779.97 MIN: 781.04 MIN: 779.2 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
TensorFlow Device: CPU - Batch Size: 512 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 512 - Model: AlexNet a b bb 50 100 150 200 250 227.16 227.85 227.00
PostgreSQL Scaling Factor: 1 - Clients: 50 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 1 - Clients: 50 - Mode: Read Write a b bb 200 400 600 800 1000 809 811 808 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a bb 20 40 60 80 100 100.52 100.15
FFmpeg Encoder: libx264 - Scenario: Platform OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Platform a b bb 9 18 27 36 45 38.19 38.07 38.21 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a bb 40 80 120 160 200 178.92 179.57
FFmpeg Encoder: libx265 - Scenario: Video On Demand OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Video On Demand a b bb 80 160 240 320 400 366.57 367.61 366.29 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 a b bb 110K 220K 330K 440K 550K 534883.8 536533.8 534630.5 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
FFmpeg Encoder: libx264 - Scenario: Platform OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Platform a b bb 40 80 120 160 200 198.35 198.96 198.27 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Embree Binary: Pathtracer ISPC - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer ISPC - Model: Asian Dragon Obj a b bb 8 16 24 32 40 32.50 32.56 32.44 MIN: 32.25 / MAX: 32.9 MIN: 32.27 / MAX: 32.95 MIN: 32.17 / MAX: 32.79
FFmpeg Encoder: libx265 - Scenario: Platform OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Platform a b bb 5 10 15 20 25 20.45 20.52 20.50 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
PostgreSQL Scaling Factor: 100 - Clients: 50 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 50 - Mode: Read Only a b bb 200K 400K 600K 800K 1000K 994655 991262 993660 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
uvg266 Video Input: Bosphorus 4K - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium a b bb 2 4 6 8 10 8.83 8.82 8.85
FFmpeg Encoder: libx265 - Scenario: Video On Demand OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Video On Demand a b bb 5 10 15 20 25 20.66 20.61 20.68 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 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream a bb 200 400 600 800 1000 965.21 961.97
RocksDB Test: Random Read OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Random Read a bb 20M 40M 60M 80M 100M 115209715 114830146 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
TensorFlow Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 a b bb 5 10 15 20 25 18.25 18.31 18.30
Google Draco Model: Church Facade OpenBenchmarking.org ms, Fewer Is Better Google Draco 1.5.6 Model: Church Facade a bb 2K 4K 6K 8K 10K 9410 9381 1. (CXX) g++ options: -O3
FFmpeg Encoder: libx265 - Scenario: Platform OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Platform a b bb 80 160 240 320 400 370.34 369.22 369.54 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
nginx Connections: 100 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 100 a bb 30K 60K 90K 120K 150K 152180.63 151733.73 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenSSL Algorithm: AES-256-GCM OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-256-GCM a b bb 30000M 60000M 90000M 120000M 150000M 118250262800 118596420950 118499993460 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Google Draco Model: Lion OpenBenchmarking.org ms, Fewer Is Better Google Draco 1.5.6 Model: Lion a bb 1400 2800 4200 5600 7000 6694 6713 1. (CXX) g++ options: -O3
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 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a bb 2 4 6 8 10 8.6680 8.6925
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 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a bb 30 60 90 120 150 115.35 115.03
Neural Magic DeepSparse Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream a bb 2 4 6 8 10 8.8732 8.8977
Neural Magic DeepSparse Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream a bb 30 60 90 120 150 112.54 112.23
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 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream a bb 20 40 60 80 100 99.49 99.75
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 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream a bb 40 80 120 160 200 180.72 180.28
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Pabellon Barcelona - Compute: CPU-Only a bb 40 80 120 160 200 191.67 192.11
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Barbershop - Compute: CPU-Only a bb 130 260 390 520 650 579.52 580.85
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU a b bb 0.1268 0.2536 0.3804 0.5072 0.634 0.563515 0.562744 0.562273 MIN: 0.54 MIN: 0.54 MIN: 0.54 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Fishy Cat - Compute: CPU-Only a bb 20 40 60 80 100 74.47 74.63
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 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a bb 30 60 90 120 150 115.16 115.41
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 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a bb 2 4 6 8 10 8.6827 8.6642
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU a b bb 2 4 6 8 10 8.67071 8.68912 8.67211 MIN: 8.58 MIN: 8.58 MIN: 8.57 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 2.12 Device: CPU - Batch Size: 64 - Model: GoogLeNet a b bb 20 40 60 80 100 77.76 77.64 77.79
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU a b bb 0.6173 1.2346 1.8519 2.4692 3.0865 2.74347 2.73821 2.73830 MIN: 2.72 MIN: 2.72 MIN: 2.72 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
John The Ripper Test: bcrypt OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: bcrypt a b bb 10K 20K 30K 40K 50K 48821 48864 48912 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lrt -lz -ldl -lcrypt -lbz2
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Classroom - Compute: CPU-Only a bb 40 80 120 160 200 159.56 159.27
Timed Node.js Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 19.8.1 Time To Compile a b bb 70 140 210 280 350 309.57 309.31 309.09
TensorFlow Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 a b bb 4 8 12 16 20 15.11 15.10 15.09
oneDNN Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU a b bb 1.281 2.562 3.843 5.124 6.405 5.69353 5.68728 5.68651 MIN: 5.55 MIN: 5.54 MIN: 5.55 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.1 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU a b bb 2 4 6 8 10 6.35319 6.35868 6.35249 MIN: 6.3 MIN: 6.3 MIN: 6.3 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
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 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream a bb 40 80 120 160 200 196.56 196.74
OpenSSL Algorithm: ChaCha20 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20 a b bb 30000M 60000M 90000M 120000M 150000M 150176760350 150074005090 150059242840 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a bb 80 160 240 320 400 351.83 352.04
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 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream a bb 20 40 60 80 100 91.41 91.35
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a bb 12 24 36 48 60 51.11 51.09
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream a bb 30 60 90 120 150 139.56 139.50
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream a bb 30 60 90 120 150 128.78 128.81
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream a bb 12 24 36 48 60 54.33 54.33
OpenSSL Algorithm: ChaCha20-Poly1305 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20-Poly1305 a b bb 20000M 40000M 60000M 80000M 100000M 80516466550 80512557330 80513755000 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream a bb 5 10 15 20 25 18.39 18.39
PostgreSQL Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average Latency a b bb 0.0113 0.0226 0.0339 0.0452 0.0565 0.05 0.05 0.05 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average Latency a b bb 0.0106 0.0212 0.0318 0.0424 0.053 0.047 0.047 0.047 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Phoronix Test Suite v10.8.4