new tests extra cpus Tests for a future article. Intel Xeon Gold 6346 testing with a Supermicro X12SPO-NTF v2.00 (1.2 BIOS) and astdrmfb on AlmaLinux 9.1 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2304246-NE-NEWTESTSE15&rdt&grr .
new tests extra cpus Processor Motherboard Memory Disk Graphics Monitor OS Kernel Compiler File-System Screen Resolution e f g h i j Intel Xeon Gold 6338 @ 3.20GHz (32 Cores / 64 Threads) Supermicro X12SPO-NTF v2.00 (1.2 BIOS) 8 x 64 GB DDR4-3200MT/s Samsung M393A8G40AB2-CWE 2 x 1920GB SAMSUNG MZQL21T9HCJR-00A07 astdrmfb DELL E207WFP AlmaLinux 9.1 5.14.0-162.12.1.el9_1.x86_64 (x86_64) GCC 11.3.1 20220421 ext4 1280x1024 Intel Xeon Gold 6346 @ 3.60GHz (16 Cores / 32 Threads) OpenBenchmarking.org Kernel Details - Transparent Huge Pages: always Compiler Details - --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-host-bind-now --enable-host-pie --enable-initfini-array --enable-languages=c,c++,fortran,lto --enable-link-serialization=1 --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=x86-64 --with-arch_64=x86-64-v2 --with-build-config=bootstrap-lto --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver --without-isl Processor Details - Scaling Governor: intel_cpufreq performance - CPU Microcode: 0xd000375 Python Details - Python 3.9.14 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Vulnerable: Clear buffers attempted no microcode; SMT vulnerable + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected
new tests extra cpus nwchem: C240 Buckyball deeprec: PLE - BF16 deeprec: PLE - BF16 deeprec: MMOE - BF16 deeprec: MMOE - BF16 deeprec: PLE - FP32 deeprec: PLE - FP32 tensorflow: CPU - 512 - ResNet-50 deeprec: BST - BF16 deeprec: BST - BF16 deeprec: DCNv2 - BF16 deeprec: DCNv2 - BF16 blender: Barbershop - CPU-Only deeprec: MMOE - FP32 deeprec: MMOE - FP32 deeprec: DCNv2 - FP32 deeprec: DCNv2 - FP32 deeprec: BST - FP32 deeprec: BST - FP32 tensorflow: CPU - 256 - ResNet-50 build-llvm: Unix Makefiles build-llvm: Ninja build-nodejs: Time To Compile petsc: Streams tensorflow: CPU - 512 - GoogLeNet ffmpeg: libx264 - Upload ffmpeg: libx264 - Upload blender: Pabellon Barcelona - CPU-Only mysqlslap: 8192 build-godot: Time To Compile deeprec: DLRM - BF16 deeprec: DLRM - BF16 opencv: Stitching blender: Classroom - CPU-Only ffmpeg: libx264 - Video On Demand ffmpeg: libx264 - Video On Demand ffmpeg: libx264 - Platform ffmpeg: libx264 - Platform ffmpeg: libx265 - Upload ffmpeg: libx265 - Upload ffmpeg: libx265 - Video On Demand ffmpeg: libx265 - Video On Demand ffmpeg: libx265 - Platform ffmpeg: libx265 - Platform memcached: 1:100 openssl: SHA256 openssl: ChaCha20-Poly1305 openssl: AES-256-GCM openssl: AES-128-GCM openssl: ChaCha20 openssl: SHA512 opencv: Graph API tensorflow: CPU - 256 - GoogLeNet mysqlslap: 512 build2: Time To Compile tensorflow: CPU - 512 - AlexNet vvenc: Bosphorus 4K - Fast tensorflow: CPU - 64 - ResNet-50 mysqlslap: 4096 mysqlslap: 2048 mysqlslap: 1024 opencv: Image Processing mysqlslap: 256 blender: Fishy Cat - CPU-Only deeprec: DLRM - FP32 deeprec: DLRM - FP32 nginx: 1000 nginx: 500 nginx: 200 apache: 1000 apache: 500 nginx: 100 apache: 200 apache: 100 stress-ng: Atomic stress-ng: NUMA stress-ng: MEMFD onednn: Recurrent Neural Network Training - u8s8f32 - CPU onednn: Recurrent Neural Network Training - f32 - CPU onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU blender: BMW27 - CPU-Only onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Recurrent Neural Network Inference - u8s8f32 - CPU stress-ng: Futex stress-ng: Socket Activity stress-ng: MEMFD opencv: Core stress-ng: Futex opencv: Object Detection tensorflow: CPU - 256 - AlexNet tensorflow: CPU - 32 - ResNet-50 memcached: 1:10 daphne: OpenMP - Points2Image memcached: 1:5 vvenc: Bosphorus 4K - Faster stress-ng: CPU Cache ffmpeg: libx265 - Live ffmpeg: libx265 - Live build-linux-kernel: defconfig john-the-ripper: MD5 rocksdb: Rand Fill rocksdb: Update Rand rocksdb: Rand Fill Sync rocksdb: Read Rand Write Rand rocksdb: Read While Writing rocksdb: Rand Read john-the-ripper: HMAC-SHA512 openssl: RSA4096 openssl: RSA4096 encode-opus: WAV To Opus Encode opencv: Features 2D svt-av1: Preset 4 - Bosphorus 4K deeprec: DIN - BF16 deeprec: DIN - BF16 tensorflow: CPU - 64 - GoogLeNet vvenc: Bosphorus 1080p - Fast ffmpeg: libx264 - Live ffmpeg: libx264 - Live deeprec: DIN - FP32 deeprec: DIN - FP32 tensorflow: CPU - 16 - ResNet-50 compress-7zip: Decompression Rating compress-7zip: Compression Rating embree: Pathtracer - Asian Dragon Obj opencv: DNN - Deep Neural Network stress-ng: Context Switching embree: Pathtracer ISPC - Asian Dragon Obj build-ffmpeg: Time To Compile rocksdb: Seq Fill stress-ng: Zlib stress-ng: Forking stress-ng: Pthread stress-ng: Hash stress-ng: Semaphores stress-ng: CPU Stress stress-ng: MMAP stress-ng: Vector Math stress-ng: Matrix Math stress-ng: Malloc stress-ng: CPU Cache stress-ng: Poll stress-ng: Crypto stress-ng: Mutex stress-ng: SENDFILE stress-ng: Zlib daphne: OpenMP - Euclidean Cluster stress-ng: Atomic john-the-ripper: WPA PSK stress-ng: Pthread stress-ng: NUMA stress-ng: Forking stress-ng: System V Message Passing stress-ng: Semaphores john-the-ripper: bcrypt john-the-ripper: Blowfish stress-ng: Vector Math stress-ng: Hash stress-ng: SENDFILE stress-ng: CPU Stress stress-ng: Malloc stress-ng: Memory Copying stress-ng: Glibc Qsort Data Sorting stress-ng: Glibc C String Functions stress-ng: Matrix Math stress-ng: MMAP stress-ng: Function Call stress-ng: Mutex stress-ng: Poll stress-ng: Crypto stress-ng: x86_64 RdRand embree: Pathtracer - Crown tensorflow: CPU - 32 - GoogLeNet embree: Pathtracer ISPC - Crown embree: Pathtracer - Asian Dragon vvenc: Bosphorus 1080p - Faster onednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPU onednn: Deconvolution Batch shapes_1d - f32 - CPU onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU tensorflow: CPU - 64 - AlexNet svt-av1: Preset 4 - Bosphorus 1080p embree: Pathtracer ISPC - Asian Dragon daphne: OpenMP - NDT Mapping onednn: IP Shapes 1D - f32 - CPU onednn: IP Shapes 1D - bf16bf16bf16 - CPU onednn: IP Shapes 1D - u8s8f32 - CPU tensorflow: CPU - 16 - GoogLeNet svt-av1: Preset 8 - Bosphorus 4K tensorflow: CPU - 32 - AlexNet onednn: IP Shapes 3D - bf16bf16bf16 - CPU onednn: IP Shapes 3D - f32 - CPU onednn: IP Shapes 3D - u8s8f32 - CPU tensorflow: CPU - 16 - AlexNet draco: Church Facade opencv: Video draco: Lion onednn: Convolution Batch Shapes Auto - f32 - CPU onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU onednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPU svt-av1: Preset 8 - Bosphorus 1080p svt-av1: Preset 12 - Bosphorus 4K svt-av1: Preset 13 - Bosphorus 4K onednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPU onednn: Deconvolution Batch shapes_3d - f32 - CPU onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU svt-av1: Preset 12 - Bosphorus 1080p svt-av1: Preset 13 - Bosphorus 1080p specfem3d: Layered Halfspace e f g h i j 4764.8 72.87 606.4 72.72 403.978 335.322 295.654 218.09 10.20 247.58 194.87 393 187.577 241748 164.85 39.48 191.89 39.33 192.59 20.64 122.34 41.74 181.488955095 42.01 180.30 2973860.21 20773085880 111222988250 267443782180 303640444380 152214168100 8346380060 188005 217.05 868 96.837 578.04 4.747 71.5 798 848 867 123678 878 76.55 164243.2 160858.82 160594.88 185846.03 186259.01 151096.19 189134.12 128779.98 156.11 179.61 430.06 903.519 886.91 891.699 59.84 513.735 511.49 508.746 2073056.27 9732.02 394.99 74952 2103532.73 30587 566.19 69.44 4268751.71 13099.407033508 4317194.51 9.046 38.99 113.16 44.63 53.833 4127000 1111179 801697 300473 2781003 4691218 106366292 147448000 419714.9 13789.3 63156 3.049 213.19 13.339 178.00 28.37 62.63 136996 196121 32.2344 29168 5166885.12 37.9247 28.014 1274287 1789.13 75346.36 142321.26 4835927.28 6279831.97 54460.73 729.26 119392.91 121611.76 78610054.56 2223417.17 4038384 39091.08 969776.95 492399.2 1620.6 690.33 157.05 189308 144344.18 166.99 77939.2 6895520.28 6280163.34 40440 40396 119310.97 4466781.04 489496.07 52534.22 78931999.99 6083.67 276.83 3501538.25 113551.29 774.15 152594.25 969913.81 4035573.84 38963.93 334310.64 29.4092 204.46 35.0031 35.4706 25.439 7.72059 4.53853 0.414775 468.66 7.231 44.4043 1165.73 1.30839 3.1127 1.59894 191.59 55.794 393.95 1.68355 1.63006 0.536328 301.94 6654 8408 5472 1.84922 1.63351 5.61518 127.605 191.683 176.746 7.77204 1.88199 0.412985 502.718 480.119 4753.6 72.89 596.97 72.78 405.544 336.192 295.08 218.2 10.20 247.53 194.12 393 187.047 239683 160.31 39.45 192.03 39.48 191.85 20.57 122.76 41.88 180.88 41.85 181.017315444 2808783.97 20783323840 111171056710 267503957880 303282814160 152207713080 8348167710 188565 217.01 873 96.64 576.64 4.755 71.49 803 853 851 122656 870 76.93 164831.27 160487.19 160596.18 185721.16 186542.92 151139.57 190175.9 129247.95 160.27 169.83 437.85 887.466 897.839 892.283 60.17 522.147 510.863 509.809 2082786.91 9783.02 415.33 76273 2125283.93 30347 565.67 69.42 4165587.24 12910.639012078 4295820.03 9.036 40.44 113.11 44.65 53.207 4119000 1109994 800319 300713 2778901 4721978 106552921 146567000 419839 13782.4 61792 3.049 213.21 13.281 178.65 28.27 62.84 136440 197139 32.0676 29288 5115907.38 38.0966 27.872 1267556 1788.2 74817.2 142232.06 4835657.84 6288116.39 54403.04 727.52 119349.95 121882.86 78868686.9 2431294.87 4035027.51 39071.55 972025.68 493574.13 1591.75 712.93 157.76 189854 144244.5 163.39 77970.3 6905447.59 6274228.79 40421 40460 119310.27 4463109.62 490342.61 52157.84 79082149.28 6066.73 275.58 3529363.29 113258.03 742.73 152593.8 969633.43 4027385.59 38948.07 334324.2 29.2865 204.59 34.9532 35.3819 25.498 7.73875 4.31228 0.414756 466.03 7.976 44.4075 1166.69 1.30087 4.19437 0.831823 191.96 56.242 393.42 1.64727 1.61819 0.560392 302 6656 8162 5492 1.69478 1.54579 5.62639 126.881 193.997 174.649 7.77293 1.86105 0.408712 514.742 483.353 10.19 247.87 391 39.37 192.41 39.47 191.89 20.61 122.50 41.75 181.45 41.85 181.01 20850017520 111103766800 266810352160 303087404370 152210202620 8346418580 866 810 849 864 887 2109816.63 9861.35 464.42 13009.342876409 40.12 112.77 44.78 53.157 4122000 1104355 797970 301369 2776570 4799172 107867894 147194000 419763.7 13813.4 3.054 178.46 28.30 136419 198114 32.1962 5204045.07 38.0698 28.036 1282869 1321.53 722.67 160.15 189308 144040.02 184.83 77787.38 7787657.63 6274788.61 40460 40402 119385.68 4460696.99 492918.31 52459.63 79143065.74 6092.1 276.13 3645309.63 113380.23 735.31 152594.67 969577.51 4023520.12 39002.26 334325.64 29.3588 34.7406 35.5055 7.515 44.2848 1142.64 55.299 6530 5525 128.142 193.389 173.494 521.86 486.252 5851.2 34.03 17420.77 60.74 31100.06 70.61 36149.21 54.14 12.35 6323.73 24.10 12340.07 889.41 188.01 96261.94 33.87 17343.97 21.31 10910.22 53.26 527.021 472.555 406.289 137247.8622 162.44 11.68 216.25 287.69 444 247.789 97.29 49814.18 202349 232.06 44.94 168.54 44.91 168.67 22.29 113.28 45.65 165.96 45.67 165.88 2136467.21 14535141250 75252309653 183035127033 207994134983 104988951623 5732289320 161540 159.46 880 107.672 400.98 4.566 52.77 826 809 829 104675 890 112.07 268.91 137681.75 109256.71 113684.79 118060.34 168568.79 168427.66 118589.58 173426.91 148306.00 260.09 95.60 538.85 1264.55 1269.70 1268.13 86.69 705.247 706.980 700.363 2969099.97 3522.26 557.29 67526 2977539.72 24753 386.26 52.15 3134084.27 19476.084982701 3380951.40 9.264 57.06 124.54 40.55 67.752 2846333 1284377 710432 211679 2353603 3122070 72619868 126293000 286665.8 9773.2 34.261 52689 3.081 82.25 42109.68 157.24 13.853 202.38 24.95 102.19 52319.06 50.52 93752 136235 21.0021 25656 5373285.42 24.3633 34.957 1498799 1203.50 64172.35 147732.48 3348679.10 3241988.27 37556.08 431.30 81804.45 84915.66 23508474.85 2473550.01 2257645.94 26814.16 1165638.38 338894.54 1135.63 961.40 260.37 129649 163837.03 93.25 71159.22 7943561.31 3244081.73 28013 28028 81800.43 3089184.07 337606.34 35599.20 23388075.58 4302.25 189.50 2408058.86 81986.25 433.20 104780.05 1165454.03 2257415.18 26680.47 334563.47 19.7176 154.60 22.4347 23.0695 27.793 11.2133 3.72347 0.494556 348.91 8.059 28.3108 1294.74 1.67460 4.16481 0.496400 150.35 47.308 306.79 1.75412 1.86015 0.633800 248.81 6390 6799 5167 1.91733 1.79436 8.12585 123.797 173.515 166.194 11.2926 2.53342 0.545388 540.944 541.995 5938.9 34.04 17426.65 60.81 31134.98 70.88 36290.11 54.12 12.39 6343.2 24.05 12314.63 889.89 188.21 96362.24 33.79 17300.83 21.35 10929.65 53.22 533.5 472.977 406.189 138390.8107 162.47 11.68 216.24 287.13 442 247.872 98.64 50501.6 203055 231.78 44.97 168.44 44.97 168.45 22.28 113.34 45.44 166.72 45.62 166.05 2164313.83 14540475710 75249418790 182814517660 208136443770 105003066980 5730328490 164124 159.34 864 105.928 400.98 4.566 52.82 822 796 822 100766 918 111.89 271.33 138919.17 109346.8 113549.16 118216.29 173404.22 173949.23 118462.79 182855.42 150468.57 273.44 94.09 391.51 1269.41 1271.33 1274.19 86.62 696.919 709.057 702.461 2815652.7 6574.21 454.13 68984 2583399.14 23632 385.88 52.21 3152408.51 18470.688556485 3411591.01 9.245 52.96 124.54 40.55 68.805 2852000 1271386 708341 211476 2337184 3111338 72533469 125816000 286768.1 9755.3 34.244 54913 3.061 82.32 42145.59 157.41 13.886 201.53 25.06 103.28 52878.34 50.64 93671 136308 20.9833 24592 5303334.58 24.2816 35.076 1485391 1155.2 64087.97 148250.98 3343242.27 3242574.79 37349.03 431.19 81676.83 85096.01 23474659.32 2540532.65 2258819.86 26817.83 1166455.13 340013.7 1140.67 1013.54 256.92 129731 163053.35 94.5 71274.12 7960924.71 3243813.64 28032 28003 81695.62 3087253.76 340726.6 35705.98 23517728.16 4310.23 190.1 2496277.19 82002.88 430.95 104763.64 1165955.27 2257168.81 26761.46 334550.11 19.7739 154.92 22.3596 23.0525 27.8 11.2282 3.7238 0.493467 349.1 8.296 28.2713 1247.75 1.67035 4.13338 0.497396 150.41 47.731 306.64 1.74051 1.82537 0.645101 248.15 6411 6693 5176 1.90404 1.73576 8.14498 123.112 172.35 166.277 11.2953 2.52744 0.551504 540.895 537.138 6087.3 34.1 17458.85 60.73 31093.88 70.58 36137.82 54.14 12.41 6354.09 24.1 12337.84 888.58 187.5 96002.56 33.84 17325.79 21.39 10952.96 53.25 522.275 473.739 405.753 139367.2041 162.25 11.66 216.52 287.41 443 248.554 98.15 50252.05 200536 232.13 44.85 168.88 44.95 168.50 22.46 112.44 45.49 166.53 45.53 166.39 1993539.11 14546066550 75274943420 181647411060 208344997610 104995097530 5727203610 160766 159.48 840 106.067 400.26 4.574 52.8 810 810 865 104033 903 112.28 269.03 137744.28 109558.98 113739.84 118198.09 168958.55 170495.74 118453.74 172379.56 149032.75 261.74 94.02 390.83 1269.07 1260.99 1263.73 86.71 702.077 699.08 713.052 2728277.39 6586.72 381.74 67760 2906460.44 23743 386.41 52.16 3161848.43 18237.248363967 3343846.78 9.332 57.86 124.90 40.43 68.867 2843000 1284890 706508 211657 2337608 3097322 71326607 125180000 286274.2 9754.1 34.313 52878 3.05 80.33 41131.21 157.38 13.823 200.92 25.13 103.14 52806.14 50.54 93599 136005 20.9765 26708 4864771.18 24.2779 35.025 1492605 1206.84 63405.19 148426.95 3342591.6 3244195.86 37389.38 431.8 80822.71 85213.63 23468513.73 2570371.44 2256203.86 26817.86 1165016.2 341456.23 1137.34 1014.53 306.3 129638 164013.17 99.76 70654.25 7955059.05 3244457.42 28022 28012 81778.72 3090952.6 342626.17 35839.55 23552788.5 4304.94 189.27 2386711.99 81862.66 431.59 104750.99 1165479.96 2254827.3 26729.2 334550.3 19.6193 154.99 22.3232 23.1589 27.91 11.577 3.72002 0.495215 348.54 8.258 28.4305 1246.03 1.67739 4.16095 0.49217 150.68 47.803 304.84 1.75393 1.83709 0.624801 248.49 6407 6813 5173 1.89081 1.77544 8.13586 123.271 173.551 167.433 11.3043 2.53518 0.545626 536.004 537.548 OpenBenchmarking.org
NWChem Input: C240 Buckyball OpenBenchmarking.org Seconds, Fewer Is Better NWChem 7.0.2 Input: C240 Buckyball e f h i j 1300 2600 3900 5200 6500 4764.8 4753.6 5851.2 5938.9 6087.3 1. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lcomex -lm -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2
DeepRec Model: PLE - Data Type: BF16 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: PLE - Data Type: BF16 h i j 8 16 24 32 40 SE +/- 0.03, N = 3 34.03 34.04 34.10
DeepRec Model: PLE - Data Type: BF16 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: PLE - Data Type: BF16 h i j 4K 8K 12K 16K 20K SE +/- 16.66, N = 3 17420.77 17426.65 17458.85
DeepRec Model: MMOE - Data Type: BF16 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: MMOE - Data Type: BF16 h i j 14 28 42 56 70 SE +/- 0.10, N = 3 60.74 60.81 60.73
DeepRec Model: MMOE - Data Type: BF16 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: MMOE - Data Type: BF16 h i j 7K 14K 21K 28K 35K SE +/- 51.49, N = 3 31100.06 31134.98 31093.88
DeepRec Model: PLE - Data Type: FP32 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: PLE - Data Type: FP32 h i j 16 32 48 64 80 SE +/- 0.14, N = 3 70.61 70.88 70.58
DeepRec Model: PLE - Data Type: FP32 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: PLE - Data Type: FP32 h i j 8K 16K 24K 32K 40K SE +/- 70.13, N = 3 36149.21 36290.11 36137.82
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 e f h i j 16 32 48 64 80 SE +/- 0.00, N = 3 72.87 72.89 54.14 54.12 54.14
DeepRec Model: BST - Data Type: BF16 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: BST - Data Type: BF16 h i j 3 6 9 12 15 SE +/- 0.01, N = 3 12.35 12.39 12.41
DeepRec Model: BST - Data Type: BF16 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: BST - Data Type: BF16 h i j 1400 2800 4200 5600 7000 SE +/- 2.60, N = 3 6323.73 6343.20 6354.09
DeepRec Model: DCNv2 - Data Type: BF16 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: DCNv2 - Data Type: BF16 h i j 6 12 18 24 30 SE +/- 0.02, N = 3 24.10 24.05 24.10
DeepRec Model: DCNv2 - Data Type: BF16 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: DCNv2 - Data Type: BF16 h i j 3K 6K 9K 12K 15K SE +/- 7.04, N = 3 12340.07 12314.63 12337.84
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Barbershop - Compute: CPU-Only e f h i j 200 400 600 800 1000 SE +/- 4.90, N = 3 606.40 596.97 889.41 889.89 888.58
DeepRec Model: MMOE - Data Type: FP32 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: MMOE - Data Type: FP32 h i j 40 80 120 160 200 SE +/- 0.11, N = 3 188.01 188.21 187.50
DeepRec Model: MMOE - Data Type: FP32 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: MMOE - Data Type: FP32 h i j 20K 40K 60K 80K 100K SE +/- 55.23, N = 3 96261.94 96362.24 96002.56
DeepRec Model: DCNv2 - Data Type: FP32 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: DCNv2 - Data Type: FP32 h i j 8 16 24 32 40 SE +/- 0.01, N = 3 33.87 33.79 33.84
DeepRec Model: DCNv2 - Data Type: FP32 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: DCNv2 - Data Type: FP32 h i j 4K 8K 12K 16K 20K SE +/- 4.89, N = 3 17343.97 17300.83 17325.79
DeepRec Model: BST - Data Type: FP32 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: BST - Data Type: FP32 h i j 5 10 15 20 25 SE +/- 0.06, N = 3 21.31 21.35 21.39
DeepRec Model: BST - Data Type: FP32 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: BST - Data Type: FP32 h i j 2K 4K 6K 8K 10K SE +/- 27.09, N = 3 10910.22 10929.65 10952.96
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 e f h i j 16 32 48 64 80 SE +/- 0.01, N = 3 72.72 72.78 53.26 53.22 53.25
Timed LLVM Compilation Build System: Unix Makefiles OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Unix Makefiles e f h i j 120 240 360 480 600 SE +/- 3.74, N = 3 403.98 405.54 527.02 533.50 522.28
Timed LLVM Compilation Build System: Ninja OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Ninja e f h i j 100 200 300 400 500 SE +/- 0.23, N = 3 335.32 336.19 472.56 472.98 473.74
Timed Node.js Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 19.8.1 Time To Compile e f h i j 90 180 270 360 450 SE +/- 0.10, N = 3 295.65 295.08 406.29 406.19 405.75
PETSc Test: Streams OpenBenchmarking.org MB/s, More Is Better PETSc 3.19 Test: Streams h i j 30K 60K 90K 120K 150K SE +/- 515.13, N = 3 137247.86 138390.81 139367.20 1. (CC) gcc options: -fPIC -O3 -O2 -lpthread -ludev -lm
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 e f h i j 50 100 150 200 250 SE +/- 0.01, N = 3 218.09 218.20 162.44 162.47 162.25
FFmpeg Encoder: libx264 - Scenario: Upload OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Upload e f g h i j 3 6 9 12 15 SE +/- 0.01, N = 3 10.20 10.20 10.19 11.68 11.68 11.66 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx264 - Scenario: Upload OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Upload e f g h i j 50 100 150 200 250 SE +/- 0.11, N = 3 247.58 247.53 247.87 216.25 216.24 216.52 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Pabellon Barcelona - Compute: CPU-Only e f h i j 60 120 180 240 300 SE +/- 0.12, N = 3 194.87 194.12 287.69 287.13 287.41
MariaDB Clients: 8192 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.0.1 Clients: 8192 e f g h i j 100 200 300 400 500 SE +/- 1.02, N = 3 393 393 391 444 442 443 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl
Timed Godot Game Engine Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 4.0 Time To Compile e f h i j 50 100 150 200 250 SE +/- 0.31, N = 3 187.58 187.05 247.79 247.87 248.55
DeepRec Model: DLRM - Data Type: BF16 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: DLRM - Data Type: BF16 h i j 20 40 60 80 100 SE +/- 0.19, N = 3 97.29 98.64 98.15
DeepRec Model: DLRM - Data Type: BF16 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: DLRM - Data Type: BF16 h i j 11K 22K 33K 44K 55K SE +/- 96.98, N = 3 49814.18 50501.60 50252.05
OpenCV Test: Stitching OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.7 Test: Stitching e f h i j 50K 100K 150K 200K 250K SE +/- 801.57, N = 3 241748 239683 202349 203055 200536 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Classroom - Compute: CPU-Only e f h i j 50 100 150 200 250 SE +/- 0.19, N = 3 164.85 160.31 232.06 231.78 232.13
FFmpeg Encoder: libx264 - Scenario: Video On Demand OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Video On Demand e f g h i j 10 20 30 40 50 SE +/- 0.02, N = 3 39.48 39.45 39.37 44.94 44.97 44.85 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx264 - Scenario: Video On Demand OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Video On Demand e f g h i j 40 80 120 160 200 SE +/- 0.06, N = 3 191.89 192.03 192.41 168.54 168.44 168.88 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx264 - Scenario: Platform OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Platform e f g h i j 10 20 30 40 50 SE +/- 0.03, N = 3 39.33 39.48 39.47 44.91 44.97 44.95 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx264 - Scenario: Platform OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Platform e f g h i j 40 80 120 160 200 SE +/- 0.12, N = 3 192.59 191.85 191.89 168.67 168.45 168.50 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx265 - Scenario: Upload OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Upload e f g h i j 5 10 15 20 25 SE +/- 0.03, N = 3 20.64 20.57 20.61 22.29 22.28 22.46 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx265 - Scenario: Upload OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Upload e f g h i j 30 60 90 120 150 SE +/- 0.17, N = 3 122.34 122.76 122.50 113.28 113.34 112.44 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx265 - Scenario: Video On Demand OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Video On Demand e f g h i j 10 20 30 40 50 SE +/- 0.09, N = 3 41.74 41.88 41.75 45.65 45.44 45.49 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx265 - Scenario: Video On Demand OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Video On Demand e f g h i j 40 80 120 160 200 SE +/- 0.33, N = 3 181.49 180.88 181.45 165.96 166.72 166.53 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx265 - Scenario: Platform OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Platform e f g h i j 10 20 30 40 50 SE +/- 0.07, N = 3 42.01 41.85 41.85 45.67 45.62 45.53 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx265 - Scenario: Platform OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Platform e f g h i j 40 80 120 160 200 SE +/- 0.25, N = 3 180.30 181.02 181.01 165.88 166.05 166.39 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -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 e f h i j 600K 1200K 1800K 2400K 3000K SE +/- 24953.13, N = 15 2973860.21 2808783.97 2136467.21 2164313.83 1993539.11 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenSSL Algorithm: SHA256 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA256 e f g h i j 4000M 8000M 12000M 16000M 20000M SE +/- 3599683.66, N = 3 20773085880 20783323840 20850017520 14535141250 14540475710 14546066550 1. (CC) gcc options: -pthread -m64 -O3 -ldl
OpenSSL Algorithm: ChaCha20-Poly1305 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20-Poly1305 e f g h i j 20000M 40000M 60000M 80000M 100000M SE +/- 6468540.55, N = 3 111222988250 111171056710 111103766800 75252309653 75249418790 75274943420 1. (CC) gcc options: -pthread -m64 -O3 -ldl
OpenSSL Algorithm: AES-256-GCM OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-256-GCM e f g h i j 60000M 120000M 180000M 240000M 300000M SE +/- 253599393.56, N = 3 267443782180 267503957880 266810352160 183035127033 182814517660 181647411060 1. (CC) gcc options: -pthread -m64 -O3 -ldl
OpenSSL Algorithm: AES-128-GCM OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-128-GCM e f g h i j 70000M 140000M 210000M 280000M 350000M SE +/- 20712512.35, N = 3 303640444380 303282814160 303087404370 207994134983 208136443770 208344997610 1. (CC) gcc options: -pthread -m64 -O3 -ldl
OpenSSL Algorithm: ChaCha20 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20 e f g h i j 30000M 60000M 90000M 120000M 150000M SE +/- 8688544.64, N = 3 152214168100 152207713080 152210202620 104988951623 105003066980 104995097530 1. (CC) gcc options: -pthread -m64 -O3 -ldl
OpenSSL Algorithm: SHA512 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA512 e f g h i j 2000M 4000M 6000M 8000M 10000M SE +/- 1617191.98, N = 3 8346380060 8348167710 8346418580 5732289320 5730328490 5727203610 1. (CC) gcc options: -pthread -m64 -O3 -ldl
OpenCV Test: Graph API OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.7 Test: Graph API e f h i j 40K 80K 120K 160K 200K SE +/- 164.80, N = 3 188005 188565 161540 164124 160766 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
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 e f h i j 50 100 150 200 250 SE +/- 0.02, N = 3 217.05 217.01 159.46 159.34 159.48
MariaDB Clients: 512 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.0.1 Clients: 512 e f g h i j 200 400 600 800 1000 SE +/- 9.84, N = 5 868 873 866 880 864 840 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl
Build2 Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.15 Time To Compile e f h i j 20 40 60 80 100 SE +/- 1.14, N = 5 96.84 96.64 107.67 105.93 106.07
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 e f h i j 120 240 360 480 600 SE +/- 0.19, N = 3 578.04 576.64 400.98 400.98 400.26
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 e f h i j 1.0699 2.1398 3.2097 4.2796 5.3495 SE +/- 0.001, N = 3 4.747 4.755 4.566 4.566 4.574 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
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 e f h i j 16 32 48 64 80 SE +/- 0.01, N = 3 71.50 71.49 52.77 52.82 52.80
MariaDB Clients: 4096 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.0.1 Clients: 4096 e f g h i j 200 400 600 800 1000 SE +/- 2.45, N = 3 798 803 810 826 822 810 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl
MariaDB Clients: 2048 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.0.1 Clients: 2048 e f g h i j 200 400 600 800 1000 SE +/- 4.69, N = 3 848 853 849 809 796 810 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl
MariaDB Clients: 1024 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.0.1 Clients: 1024 e f g h i j 200 400 600 800 1000 SE +/- 8.73, N = 3 867 851 864 829 822 865 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl
OpenCV Test: Image Processing OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.7 Test: Image Processing e f h i j 30K 60K 90K 120K 150K SE +/- 916.27, N = 3 123678 122656 104675 100766 104033 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
MariaDB Clients: 256 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.0.1 Clients: 256 e f g h i j 200 400 600 800 1000 SE +/- 10.01, N = 3 878 870 887 890 918 903 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -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 e f h i j 30 60 90 120 150 SE +/- 0.09, N = 3 76.55 76.93 112.07 111.89 112.28
DeepRec Model: DLRM - Data Type: FP32 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: DLRM - Data Type: FP32 h i j 60 120 180 240 300 SE +/- 0.29, N = 3 268.91 271.33 269.03
DeepRec Model: DLRM - Data Type: FP32 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: DLRM - Data Type: FP32 h i j 30K 60K 90K 120K 150K SE +/- 145.81, N = 3 137681.75 138919.17 137744.28
nginx Connections: 1000 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 e f h i j 40K 80K 120K 160K 200K SE +/- 28.90, N = 3 164243.20 164831.27 109256.71 109346.80 109558.98 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
nginx Connections: 500 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 e f h i j 30K 60K 90K 120K 150K SE +/- 17.26, N = 3 160858.82 160487.19 113684.79 113549.16 113739.84 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
nginx Connections: 200 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 200 e f h i j 30K 60K 90K 120K 150K SE +/- 10.78, N = 3 160594.88 160596.18 118060.34 118216.29 118198.09 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Apache HTTP Server Concurrent Requests: 1000 OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.56 Concurrent Requests: 1000 e f h i j 40K 80K 120K 160K 200K SE +/- 303.73, N = 3 185846.03 185721.16 168568.79 173404.22 168958.55 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Apache HTTP Server Concurrent Requests: 500 OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.56 Concurrent Requests: 500 e f h i j 40K 80K 120K 160K 200K SE +/- 306.33, N = 3 186259.01 186542.92 168427.66 173949.23 170495.74 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
nginx Connections: 100 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 100 e f h i j 30K 60K 90K 120K 150K SE +/- 33.69, N = 3 151096.19 151139.57 118589.58 118462.79 118453.74 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Apache HTTP Server Concurrent Requests: 200 OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.56 Concurrent Requests: 200 e f h i j 40K 80K 120K 160K 200K SE +/- 387.27, N = 3 189134.12 190175.90 173426.91 182855.42 172379.56 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Apache HTTP Server Concurrent Requests: 100 OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.56 Concurrent Requests: 100 e f h i j 30K 60K 90K 120K 150K SE +/- 69.71, N = 3 128779.98 129247.95 148306.00 150468.57 149032.75 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Stress-NG Test: Atomic OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Atomic e f h i j 60 120 180 240 300 SE +/- 2.20, N = 15 156.11 160.27 260.09 273.44 261.74 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: NUMA OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: NUMA e f h i j 40 80 120 160 200 SE +/- 1.20, N = 15 179.61 169.83 95.60 94.09 94.02 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: MEMFD OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: MEMFD e f h i j 120 240 360 480 600 SE +/- 14.06, N = 15 430.06 437.85 538.85 391.51 390.83 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
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 e f h i j 300 600 900 1200 1500 SE +/- 2.92, N = 3 903.52 887.47 1264.55 1269.41 1269.07 MIN: 849.4 MIN: 844.57 MIN: 1210.64 MIN: 1219.37 MIN: 1220.45 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 300 600 900 1200 1500 SE +/- 2.73, N = 3 886.91 897.84 1269.70 1271.33 1260.99 MIN: 844.74 MIN: 851.56 MIN: 1215.95 MIN: 1221.68 MIN: 1212.22 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 300 600 900 1200 1500 SE +/- 3.00, N = 3 891.70 892.28 1268.13 1274.19 1263.73 MIN: 849.48 MIN: 849.32 MIN: 1214.46 MIN: 1225.59 MIN: 1212.79 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: BMW27 - Compute: CPU-Only e f h i j 20 40 60 80 100 SE +/- 0.02, N = 3 59.84 60.17 86.69 86.62 86.71
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 e f h i j 150 300 450 600 750 SE +/- 0.94, N = 3 513.74 522.15 705.25 696.92 702.08 MIN: 495.91 MIN: 502.11 MIN: 674.1 MIN: 668.26 MIN: 673.18 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 150 300 450 600 750 SE +/- 0.30, N = 3 511.49 510.86 706.98 709.06 699.08 MIN: 490.8 MIN: 492.47 MIN: 677.17 MIN: 677.09 MIN: 668.9 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 150 300 450 600 750 SE +/- 2.34, N = 3 508.75 509.81 700.36 702.46 713.05 MIN: 489.62 MIN: 489.85 MIN: 667.01 MIN: 672.81 MIN: 680.59 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Stress-NG Test: Futex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Futex e f g h i j 600K 1200K 1800K 2400K 3000K SE +/- 44269.26, N = 15 2073056.27 2082786.91 2109816.63 2969099.97 2815652.70 2728277.39 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Socket Activity OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Socket Activity e f g h i j 2K 4K 6K 8K 10K SE +/- 622.10, N = 15 9732.02 9783.02 9861.35 3522.26 6574.21 6586.72 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: MEMFD OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: MEMFD e f g h i j 120 240 360 480 600 SE +/- 13.12, N = 15 394.99 415.33 464.42 557.29 454.13 381.74 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
OpenCV Test: Core OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.7 Test: Core e f h i j 16K 32K 48K 64K 80K SE +/- 723.19, N = 3 74952 76273 67526 68984 67760 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
Stress-NG Test: Futex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Futex e f h i j 600K 1200K 1800K 2400K 3000K SE +/- 51354.36, N = 12 2103532.73 2125283.93 2977539.72 2583399.14 2906460.44 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
OpenCV Test: Object Detection OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.7 Test: Object Detection e f h i j 7K 14K 21K 28K 35K SE +/- 286.60, N = 15 30587 30347 24753 23632 23743 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
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 e f h i j 120 240 360 480 600 SE +/- 0.07, N = 3 566.19 565.67 386.26 385.88 386.41
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 e f h i j 15 30 45 60 75 SE +/- 0.01, N = 3 69.44 69.42 52.15 52.21 52.16
Memcached Set To Get Ratio: 1:10 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:10 e f h i j 900K 1800K 2700K 3600K 4500K SE +/- 26057.18, N = 3 4268751.71 4165587.24 3134084.27 3152408.51 3161848.43 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Points2Image OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite 2021.11.02 Backend: OpenMP - Kernel: Points2Image e f g h i j 4K 8K 12K 16K 20K SE +/- 30.76, N = 3 13099.41 12910.64 13009.34 19476.08 18470.69 18237.25 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
Memcached Set To Get Ratio: 1:5 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:5 e f h i j 900K 1800K 2700K 3600K 4500K SE +/- 12477.68, N = 3 4317194.51 4295820.03 3380951.40 3411591.01 3343846.78 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
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 e f h i j 3 6 9 12 15 SE +/- 0.012, N = 3 9.046 9.036 9.264 9.245 9.332 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
Stress-NG Test: CPU Cache OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: CPU Cache e f g h i j 13 26 39 52 65 SE +/- 3.29, N = 12 38.99 40.44 40.12 57.06 52.96 57.86 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
FFmpeg Encoder: libx265 - Scenario: Live OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Live e f g h i j 30 60 90 120 150 SE +/- 0.19, N = 3 113.16 113.11 112.77 124.54 124.54 124.90 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx265 - Scenario: Live OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx265 - Scenario: Live e f g h i j 10 20 30 40 50 SE +/- 0.06, N = 3 44.63 44.65 44.78 40.55 40.55 40.43 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
Timed Linux Kernel Compilation Build: defconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: defconfig e f g h i j 15 30 45 60 75 SE +/- 0.52, N = 3 53.83 53.21 53.16 67.75 68.81 68.87
John The Ripper Test: MD5 OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: MD5 e f g h i j 900K 1800K 2700K 3600K 4500K SE +/- 2962.73, N = 3 4127000 4119000 4122000 2846333 2852000 2843000 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt
RocksDB Test: Random Fill OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Random Fill e f g h i j 300K 600K 900K 1200K 1500K SE +/- 4097.59, N = 3 1111179 1109994 1104355 1284377 1271386 1284890 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti
RocksDB Test: Update Random OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Update Random e f g h i j 200K 400K 600K 800K 1000K SE +/- 188.07, N = 3 801697 800319 797970 710432 708341 706508 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti
RocksDB Test: Random Fill Sync OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Random Fill Sync e f g h i j 60K 120K 180K 240K 300K SE +/- 58.27, N = 3 300473 300713 301369 211679 211476 211657 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti
RocksDB Test: Read Random Write Random OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Read Random Write Random e f g h i j 600K 1200K 1800K 2400K 3000K SE +/- 6073.00, N = 3 2781003 2778901 2776570 2353603 2337184 2337608 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti
RocksDB Test: Read While Writing OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Read While Writing e f g h i j 1000K 2000K 3000K 4000K 5000K SE +/- 10221.65, N = 3 4691218 4721978 4799172 3122070 3111338 3097322 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti
RocksDB Test: Random Read OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Random Read e f g h i j 20M 40M 60M 80M 100M SE +/- 707833.96, N = 3 106366292 106552921 107867894 72619868 72533469 71326607 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti
John The Ripper Test: HMAC-SHA512 OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: HMAC-SHA512 e f g h i j 30M 60M 90M 120M 150M SE +/- 405431.05, N = 3 147448000 146567000 147194000 126293000 125816000 125180000 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 e f g h i j 90K 180K 270K 360K 450K SE +/- 136.10, N = 3 419714.9 419839.0 419763.7 286665.8 286768.1 286274.2 1. (CC) gcc options: -pthread -m64 -O3 -ldl
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 e f g h i j 3K 6K 9K 12K 15K SE +/- 2.48, N = 3 13789.3 13782.4 13813.4 9773.2 9755.3 9754.1 1. (CC) gcc options: -pthread -m64 -O3 -ldl
Opus Codec Encoding WAV To Opus Encode OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.4 WAV To Opus Encode h i j 8 16 24 32 40 SE +/- 0.01, N = 5 34.26 34.24 34.31 1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm
OpenCV Test: Features 2D OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.7 Test: Features 2D e f h i j 14K 28K 42K 56K 70K SE +/- 672.84, N = 3 63156 61792 52689 54913 52878 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
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 e f g h i j 0.6932 1.3864 2.0796 2.7728 3.466 SE +/- 0.015, N = 3 3.049 3.049 3.054 3.081 3.061 3.050 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
DeepRec Model: DIN - Data Type: BF16 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: DIN - Data Type: BF16 h i j 20 40 60 80 100 SE +/- 0.13, N = 3 82.25 82.32 80.33
DeepRec Model: DIN - Data Type: BF16 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: DIN - Data Type: BF16 h i j 9K 18K 27K 36K 45K SE +/- 67.58, N = 3 42109.68 42145.59 41131.21
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 e f h i j 50 100 150 200 250 SE +/- 0.13, N = 3 213.19 213.21 157.24 157.41 157.38
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 e f h i j 4 8 12 16 20 SE +/- 0.04, N = 3 13.34 13.28 13.85 13.89 13.82 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
FFmpeg Encoder: libx264 - Scenario: Live OpenBenchmarking.org FPS, More Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Live e f g h i j 40 80 120 160 200 SE +/- 0.33, N = 3 178.00 178.65 178.46 202.38 201.53 200.92 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
FFmpeg Encoder: libx264 - Scenario: Live OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 6.0 Encoder: libx264 - Scenario: Live e f g h i j 7 14 21 28 35 SE +/- 0.04, N = 3 28.37 28.27 28.30 24.95 25.06 25.13 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl
DeepRec Model: DIN - Data Type: FP32 OpenBenchmarking.org Gstep / sec, More Is Better DeepRec Model: DIN - Data Type: FP32 h i j 20 40 60 80 100 SE +/- 0.15, N = 3 102.19 103.28 103.14
DeepRec Model: DIN - Data Type: FP32 OpenBenchmarking.org Throughput, More Is Better DeepRec Model: DIN - Data Type: FP32 h i j 11K 22K 33K 44K 55K SE +/- 79.54, N = 3 52319.06 52878.34 52806.14
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 e f h i j 14 28 42 56 70 SE +/- 0.02, N = 3 62.63 62.84 50.52 50.64 50.54
7-Zip Compression Test: Decompression Rating OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating e f g h i j 30K 60K 90K 120K 150K SE +/- 74.28, N = 3 136996 136440 136419 93752 93671 93599 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
7-Zip Compression Test: Compression Rating OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating e f g h i j 40K 80K 120K 160K 200K SE +/- 426.89, N = 3 196121 197139 198114 136235 136308 136005 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
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 e f g h i j 7 14 21 28 35 SE +/- 0.04, N = 3 32.23 32.07 32.20 21.00 20.98 20.98 MIN: 31.25 / MAX: 33.72 MIN: 31.19 / MAX: 33.77 MIN: 31.14 / MAX: 33.69 MIN: 20.84 / MAX: 21.33 MIN: 20.9 / MAX: 21.18 MIN: 20.89 / MAX: 21.22
OpenCV Test: DNN - Deep Neural Network OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.7 Test: DNN - Deep Neural Network e f h i j 6K 12K 18K 24K 30K SE +/- 264.62, N = 5 29168 29288 25656 24592 26708 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
Stress-NG Test: Context Switching OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Context Switching e f g h i j 1.2M 2.4M 3.6M 4.8M 6M SE +/- 61475.84, N = 4 5166885.12 5115907.38 5204045.07 5373285.42 5303334.58 4864771.18 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
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 e f g h i j 9 18 27 36 45 SE +/- 0.03, N = 3 37.92 38.10 38.07 24.36 24.28 24.28 MIN: 37.43 / MAX: 40.04 MIN: 37.56 / MAX: 40.12 MIN: 37.55 / MAX: 40.11 MIN: 23.83 / MAX: 26.05 MIN: 23.75 / MAX: 25.83 MIN: 23.75 / MAX: 25.58
Timed FFmpeg Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed FFmpeg Compilation 6.0 Time To Compile e f g h i j 8 16 24 32 40 SE +/- 0.03, N = 3 28.01 27.87 28.04 34.96 35.08 35.03
RocksDB Test: Sequential Fill OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Sequential Fill e f g h i j 300K 600K 900K 1200K 1500K SE +/- 5575.25, N = 3 1274287 1267556 1282869 1498799 1485391 1492605 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti
Stress-NG Test: Zlib OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Zlib e f h i j 400 800 1200 1600 2000 SE +/- 1.64, N = 3 1789.13 1788.20 1203.50 1155.20 1206.84 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Forking OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Forking e f h i j 16K 32K 48K 64K 80K SE +/- 221.26, N = 3 75346.36 74817.20 64172.35 64087.97 63405.19 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Pthread OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Pthread e f h i j 30K 60K 90K 120K 150K SE +/- 671.65, N = 3 142321.26 142232.06 147732.48 148250.98 148426.95 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Hash OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Hash e f h i j 1000K 2000K 3000K 4000K 5000K SE +/- 1653.89, N = 3 4835927.28 4835657.84 3348679.10 3343242.27 3342591.60 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Semaphores e f h i j 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 1059.44, N = 3 6279831.97 6288116.39 3241988.27 3242574.79 3244195.86 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: CPU Stress OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: CPU Stress e f h i j 12K 24K 36K 48K 60K SE +/- 54.92, N = 3 54460.73 54403.04 37556.08 37349.03 37389.38 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: MMAP OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: MMAP e f h i j 160 320 480 640 800 SE +/- 0.39, N = 3 729.26 727.52 431.30 431.19 431.80 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Vector Math e f h i j 30K 60K 90K 120K 150K SE +/- 56.68, N = 3 119392.91 119349.95 81804.45 81676.83 80822.71 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Matrix Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Matrix Math e f h i j 30K 60K 90K 120K 150K SE +/- 51.94, N = 3 121611.76 121882.86 84915.66 85096.01 85213.63 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Malloc OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Malloc e f h i j 20M 40M 60M 80M 100M SE +/- 12210.94, N = 3 78610054.56 78868686.90 23508474.85 23474659.32 23468513.73 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: CPU Cache OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: CPU Cache e f h i j 600K 1200K 1800K 2400K 3000K SE +/- 35357.11, N = 3 2223417.17 2431294.87 2473550.01 2540532.65 2570371.44 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Poll OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Poll e f h i j 900K 1800K 2700K 3600K 4500K SE +/- 1054.76, N = 3 4038384.00 4035027.51 2257645.94 2258819.86 2256203.86 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Crypto OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Crypto e f h i j 8K 16K 24K 32K 40K SE +/- 2.51, N = 3 39091.08 39071.55 26814.16 26817.83 26817.86 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Mutex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: Mutex e f h i j 200K 400K 600K 800K 1000K SE +/- 265.04, N = 3 969776.95 972025.68 1165638.38 1166455.13 1165016.20 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: SENDFILE OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.06 Test: SENDFILE e f h i j 110K 220K 330K 440K 550K SE +/- 82.32, N = 3 492399.20 493574.13 338894.54 340013.70 341456.23 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
Stress-NG Test: Zlib OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Zlib e f g h i j 300 600 900 1200 1500 SE +/- 1.93, N = 3 1620.60 1591.75 1321.53 1135.63 1140.67 1137.34 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Euclidean Cluster OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite 2021.11.02 Backend: OpenMP - Kernel: Euclidean Cluster e f g h i j 200 400 600 800 1000 SE +/- 10.31, N = 5 690.33 712.93 722.67 961.40 1013.54 1014.53 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
Stress-NG Test: Atomic OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Atomic e f g h i j 70 140 210 280 350 SE +/- 1.08, N = 3 157.05 157.76 160.15 260.37 256.92 306.30 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
John The Ripper Test: WPA PSK OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: WPA PSK e f g h i j 40K 80K 120K 160K 200K SE +/- 27.79, N = 3 189308 189854 189308 129649 129731 129638 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt
Stress-NG Test: Pthread OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Pthread e f g h i j 40K 80K 120K 160K 200K SE +/- 150.68, N = 3 144344.18 144244.50 144040.02 163837.03 163053.35 164013.17 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: NUMA OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: NUMA e f g h i j 40 80 120 160 200 SE +/- 0.91, N = 3 166.99 163.39 184.83 93.25 94.50 99.76 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Forking OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Forking e f g h i j 20K 40K 60K 80K 100K SE +/- 132.55, N = 3 77939.20 77970.30 77787.38 71159.22 71274.12 70654.25 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: System V Message Passing OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: System V Message Passing e f g h i j 2M 4M 6M 8M 10M SE +/- 4165.71, N = 3 6895520.28 6905447.59 7787657.63 7943561.31 7960924.71 7955059.05 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Semaphores e f g h i j 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 626.84, N = 3 6280163.34 6274228.79 6274788.61 3244081.73 3243813.64 3244457.42 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
John The Ripper Test: bcrypt OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: bcrypt e f g h i j 9K 18K 27K 36K 45K SE +/- 11.26, N = 3 40440 40421 40460 28013 28032 28022 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: Blowfish e f g h i j 9K 18K 27K 36K 45K SE +/- 11.70, N = 3 40396 40460 40402 28028 28003 28012 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt
Stress-NG Test: Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Vector Math e f g h i j 30K 60K 90K 120K 150K SE +/- 32.35, N = 3 119310.97 119310.27 119385.68 81800.43 81695.62 81778.72 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Hash OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Hash e f g h i j 1000K 2000K 3000K 4000K 5000K SE +/- 126.06, N = 3 4466781.04 4463109.62 4460696.99 3089184.07 3087253.76 3090952.60 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: SENDFILE OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: SENDFILE e f g h i j 110K 220K 330K 440K 550K SE +/- 71.80, N = 3 489496.07 490342.61 492918.31 337606.34 340726.60 342626.17 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: CPU Stress OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: CPU Stress e f g h i j 11K 22K 33K 44K 55K SE +/- 35.16, N = 3 52534.22 52157.84 52459.63 35599.20 35705.98 35839.55 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Malloc OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Malloc e f g h i j 20M 40M 60M 80M 100M SE +/- 109768.81, N = 3 78931999.99 79082149.28 79143065.74 23388075.58 23517728.16 23552788.50 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Memory Copying OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Memory Copying e f g h i j 1300 2600 3900 5200 6500 SE +/- 0.91, N = 3 6083.67 6066.73 6092.10 4302.25 4310.23 4304.94 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Glibc Qsort Data Sorting OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Glibc Qsort Data Sorting e f g h i j 60 120 180 240 300 SE +/- 0.26, N = 3 276.83 275.58 276.13 189.50 190.10 189.27 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Glibc C String Functions OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Glibc C String Functions e f g h i j 800K 1600K 2400K 3200K 4000K SE +/- 3246.22, N = 3 3501538.25 3529363.29 3645309.63 2408058.86 2496277.19 2386711.99 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Matrix Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Matrix Math e f g h i j 20K 40K 60K 80K 100K SE +/- 62.64, N = 3 113551.29 113258.03 113380.23 81986.25 82002.88 81862.66 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: MMAP OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: MMAP e f g h i j 170 340 510 680 850 SE +/- 0.70, N = 3 774.15 742.73 735.31 433.20 430.95 431.59 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Function Call OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Function Call e f g h i j 30K 60K 90K 120K 150K SE +/- 3.58, N = 3 152594.25 152593.80 152594.67 104780.05 104763.64 104750.99 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Mutex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Mutex e f g h i j 200K 400K 600K 800K 1000K SE +/- 196.52, N = 3 969913.81 969633.43 969577.51 1165454.03 1165955.27 1165479.96 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Poll OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Poll e f g h i j 900K 1800K 2700K 3600K 4500K SE +/- 230.77, N = 3 4035573.84 4027385.59 4023520.12 2257415.18 2257168.81 2254827.30 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: Crypto OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: Crypto e f g h i j 8K 16K 24K 32K 40K SE +/- 22.35, N = 3 38963.93 38948.07 39002.26 26680.47 26761.46 26729.20 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Stress-NG Test: x86_64 RdRand OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.04 Test: x86_64 RdRand e f g h i j 70K 140K 210K 280K 350K SE +/- 10.75, N = 3 334310.64 334324.20 334325.64 334563.47 334550.11 334550.30 1. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer - Model: Crown e f g h i j 7 14 21 28 35 SE +/- 0.05, N = 3 29.41 29.29 29.36 19.72 19.77 19.62 MIN: 28.72 / MAX: 31.05 MIN: 28.61 / MAX: 30.83 MIN: 28.73 / MAX: 30.61 MIN: 19.49 / MAX: 20.5 MIN: 19.64 / MAX: 20.73 MIN: 19.47 / MAX: 20.27
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 e f h i j 40 80 120 160 200 SE +/- 0.08, N = 3 204.46 204.59 154.60 154.92 154.99
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer ISPC - Model: Crown e f g h i j 8 16 24 32 40 SE +/- 0.04, N = 3 35.00 34.95 34.74 22.43 22.36 22.32 MIN: 34.13 / MAX: 36.64 MIN: 34.07 / MAX: 36.73 MIN: 33.94 / MAX: 36.36 MIN: 21.85 / MAX: 24.32 MIN: 21.84 / MAX: 24 MIN: 21.84 / MAX: 24.02
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0.1 Binary: Pathtracer - Model: Asian Dragon e f g h i j 8 16 24 32 40 SE +/- 0.05, N = 3 35.47 35.38 35.51 23.07 23.05 23.16 MIN: 34.22 / MAX: 37.08 MIN: 34.3 / MAX: 36.96 MIN: 34.23 / MAX: 37 MIN: 22.9 / MAX: 23.77 MIN: 22.96 / MAX: 23.68 MIN: 23.06 / MAX: 23.75
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 e f h i j 7 14 21 28 35 SE +/- 0.04, N = 3 25.44 25.50 27.79 27.80 27.91 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
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 e f h i j 3 6 9 12 15 SE +/- 0.00893, N = 3 7.72059 7.73875 11.21330 11.22820 11.57700 MIN: 7.58 MIN: 7.58 MIN: 11.11 MIN: 11.11 MIN: 11.48 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 1.0212 2.0424 3.0636 4.0848 5.106 SE +/- 0.00329, N = 3 4.53853 4.31228 3.72347 3.72380 3.72002 MIN: 3.54 MIN: 3.56 MIN: 3.03 MIN: 3.04 MIN: 3.1 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 0.1114 0.2228 0.3342 0.4456 0.557 SE +/- 0.000235, N = 3 0.414775 0.414756 0.494556 0.493467 0.495215 MIN: 0.35 MIN: 0.36 MIN: 0.42 MIN: 0.42 MIN: 0.42 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -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 e f h i j 100 200 300 400 500 SE +/- 0.39, N = 3 468.66 466.03 348.91 349.10 348.54
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 e f g h i j 2 4 6 8 10 SE +/- 0.105, N = 3 7.231 7.976 7.515 8.059 8.296 8.258 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
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 e f g h i j 10 20 30 40 50 SE +/- 0.07, N = 3 44.40 44.41 44.28 28.31 28.27 28.43 MIN: 43.8 / MAX: 46.79 MIN: 43.88 / MAX: 46.85 MIN: 43.73 / MAX: 46.57 MIN: 27.61 / MAX: 30.41 MIN: 27.6 / MAX: 30.18 MIN: 27.79 / MAX: 30.42
Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: NDT Mapping OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite 2021.11.02 Backend: OpenMP - Kernel: NDT Mapping e f g h i j 300 600 900 1200 1500 SE +/- 0.51, N = 3 1165.73 1166.69 1142.64 1294.74 1247.75 1246.03 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
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 e f h i j 0.3774 0.7548 1.1322 1.5096 1.887 SE +/- 0.00645, N = 3 1.30839 1.30087 1.67460 1.67035 1.67739 MIN: 1.19 MIN: 1.17 MIN: 1.45 MIN: 1.44 MIN: 1.45 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 0.9437 1.8874 2.8311 3.7748 4.7185 SE +/- 0.01380, N = 3 3.11270 4.19437 4.16481 4.13338 4.16095 MIN: 2.91 MIN: 2.98 MIN: 4.03 MIN: 4.03 MIN: 4.03 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 0.3598 0.7196 1.0794 1.4392 1.799 SE +/- 0.002966, N = 3 1.598940 0.831823 0.496400 0.497396 0.492170 MIN: 0.67 MIN: 0.53 MIN: 0.39 MIN: 0.4 MIN: 0.39 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 40 80 120 160 200 SE +/- 0.06, N = 3 191.59 191.96 150.35 150.41 150.68
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 e f g h i j 13 26 39 52 65 SE +/- 0.12, N = 3 55.79 56.24 55.30 47.31 47.73 47.80 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
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 e f h i j 90 180 270 360 450 SE +/- 0.15, N = 3 393.95 393.42 306.79 306.64 304.84
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 e f h i j 0.3947 0.7894 1.1841 1.5788 1.9735 SE +/- 0.00448, N = 3 1.68355 1.64727 1.75412 1.74051 1.75393 MIN: 1.42 MIN: 1.42 MIN: 1.69 MIN: 1.69 MIN: 1.7 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -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 e f h i j 0.4185 0.837 1.2555 1.674 2.0925 SE +/- 0.00420, N = 3 1.63006 1.61819 1.86015 1.82537 1.83709 MIN: 1.58 MIN: 1.57 MIN: 1.79 MIN: 1.77 MIN: 1.76 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 0.1451 0.2902 0.4353 0.5804 0.7255 SE +/- 0.006251, N = 3 0.536328 0.560392 0.633800 0.645101 0.624801 MIN: 0.48 MIN: 0.51 MIN: 0.57 MIN: 0.59 MIN: 0.57 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 70 140 210 280 350 SE +/- 0.29, N = 3 301.94 302.00 248.81 248.15 248.49
Google Draco Model: Church Facade OpenBenchmarking.org ms, Fewer Is Better Google Draco 1.5.6 Model: Church Facade e f g h i j 1400 2800 4200 5600 7000 SE +/- 1.20, N = 3 6654 6656 6530 6390 6411 6407 1. (CXX) g++ options: -O3
OpenCV Test: Video OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.7 Test: Video e f h i j 2K 4K 6K 8K 10K SE +/- 13.02, N = 3 8408 8162 6799 6693 6813 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
Google Draco Model: Lion OpenBenchmarking.org ms, Fewer Is Better Google Draco 1.5.6 Model: Lion e f g h i j 1200 2400 3600 4800 6000 SE +/- 5.78, N = 3 5472 5492 5525 5167 5176 5173 1. (CXX) g++ options: -O3
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 e f h i j 0.4314 0.8628 1.2942 1.7256 2.157 SE +/- 0.00961, N = 3 1.84922 1.69478 1.91733 1.90404 1.89081 MIN: 1.79 MIN: 1.62 MIN: 1.75 MIN: 1.74 MIN: 1.73 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 0.4037 0.8074 1.2111 1.6148 2.0185 SE +/- 0.01503, N = 3 1.63351 1.54579 1.79436 1.73576 1.77544 MIN: 1.57 MIN: 1.48 MIN: 1.69 MIN: 1.64 MIN: 1.7 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -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 e f h i j 2 4 6 8 10 SE +/- 0.00039, N = 3 5.61518 5.62639 8.12585 8.14498 8.13586 MIN: 5.59 MIN: 5.59 MIN: 8.12 MIN: 8.12 MIN: 8.12 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -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 e f g h i j 30 60 90 120 150 SE +/- 0.32, N = 3 127.61 126.88 128.14 123.80 123.11 123.27 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
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 e f g h i j 40 80 120 160 200 SE +/- 0.03, N = 3 191.68 194.00 193.39 173.52 172.35 173.55 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
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 e f g h i j 40 80 120 160 200 SE +/- 0.56, N = 3 176.75 174.65 173.49 166.19 166.28 167.43 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
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 e f h i j 3 6 9 12 15 SE +/- 0.00717, N = 3 7.77204 7.77293 11.29260 11.29530 11.30430 MIN: 7.7 MIN: 7.71 MIN: 11.22 MIN: 11.22 MIN: 11.23 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 0.5704 1.1408 1.7112 2.2816 2.852 SE +/- 0.00464, N = 3 1.88199 1.86105 2.53342 2.52744 2.53518 MIN: 1.65 MIN: 1.65 MIN: 2.24 MIN: 2.31 MIN: 2.31 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f h i j 0.1241 0.2482 0.3723 0.4964 0.6205 SE +/- 0.000311, N = 3 0.412985 0.408712 0.545388 0.551504 0.545626 MIN: 0.34 MIN: 0.34 MIN: 0.45 MIN: 0.45 MIN: 0.45 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 e f g h i j 120 240 360 480 600 SE +/- 4.52, N = 3 502.72 514.74 521.86 540.94 540.90 536.00 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
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 e f g h i j 120 240 360 480 600 SE +/- 2.33, N = 3 480.12 483.35 486.25 542.00 537.14 537.55 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Phoronix Test Suite v10.8.4