september 2021 AMD Ryzen Threadripper 2950X 16-Core testing with a MSI MEG X399 CREATION (MS-7B92) v1.0 (1.30 BIOS) and AMD FirePro V4900 1GB on Debian 10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2109286-TJ-SEPTEMBER11&sor&grs .
september 2021 Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Compiler File-System Screen Resolution 1 2 2a 3 4 AMD Ryzen Threadripper 2950X 16-Core @ 3.50GHz (16 Cores / 32 Threads) MSI MEG X399 CREATION (MS-7B92) v1.0 (1.30 BIOS) AMD 17h 32GB Samsung SSD 970 EVO 250GB AMD FirePro V4900 1GB Realtek ALC1220 VA2431 2 x Intel I211 + Intel-AC 9260 Debian 10 5.8.1-050801-generic (x86_64) GNOME Shell 3.30.2 X Server 1.20.4 + Wayland 3.3 Mesa 18.3.6 (LLVM 7.0.1) GCC 8.3.0 ext4 1920x1080 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v Disk Details - 1, 2: NONE / errors=remount-ro,relatime,rw / Block Size: 4096 Processor Details - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x800820b Java Details - OpenJDK Runtime Environment (build 11.0.8+10-post-Debian-1deb10u1) Python Details - Python 2.7.16 + Python 3.7.3 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + srbds: Not affected + tsx_async_abort: Not affected
september 2021 ncnn: CPU - googlenet yafaray: Total Time For Sample Scene mrbayes: Primate Phylogeny Analysis renaissance: Savina Reactors.IO ncnn: CPU - mnasnet ncnn: CPU - regnety_400m mnn: squeezenetv1.1 ncnn: CPU-v3-v3 - mobilenet-v3 ncnn: CPU - efficientnet-b0 mnn: SqueezeNetV1.0 natron: Spaceship ncnn: CPU - alexnet mnn: mobilenetV3 compress-zstd: 8, Long Mode - Compression Speed onnx: shufflenet-v2-10 - OpenMP CPU ncnn: CPU-v2-v2 - mobilenet-v2 openssl: SHA256 compress-zstd: 8 - Compression Speed ncnn: CPU - resnet50 srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAM ncnn: CPU - yolov4-tiny srsran: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM mnn: MobileNetV2_224 compress-zstd: 19 - Compression Speed yquake2: Software CPU - 1920 x 1080 renaissance: Apache Spark PageRank stress-ng: CPU Cache embree: Pathtracer - Asian Dragon renaissance: Akka Unbalanced Cobwebbed Tree ncnn: CPU - mobilenet build-linux-kernel: Time To Compile ncnn: CPU - shufflenet-v2 hmmer: Pfam Database Search botan: CAST-256 keydb: ncnn: CPU - squeezenet_ssd stress-ng: NUMA compress-zstd: 3 - Compression Speed stress-ng: Memory Copying botan: CAST-256 - Decrypt ncnn: CPU - resnet18 build-ffmpeg: Time To Compile embree: Pathtracer ISPC - Crown embree: Pathtracer ISPC - Asian Dragon Obj botan: AES-256 - Decrypt botan: AES-256 onnx: bertsquad-10 - OpenMP CPU compress-zstd: 19 - Decompression Speed jpegxl: PNG - 8 jpegxl: JPEG - 7 srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAM build-llvm: Unix Makefiles npb: SP.B draco: Church Facade stress-ng: MMAP srsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAM stress-ng: Glibc Qsort Data Sorting mnn: inception-v3 tnn: CPU - SqueezeNet v1.1 botan: Blowfish - Decrypt vpxenc: Speed 5 - Bosphorus 1080p stress-ng: Context Switching embree: Pathtracer - Asian Dragon Obj jpegxl: JPEG - 8 srsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAM yquake2: OpenGL 3.x - 1920 x 1080 stress-ng: Glibc C String Functions synthmark: VoiceMark_100 renaissance: In-Memory Database Shootout botan: Blowfish renaissance: Finagle HTTP Requests renaissance: Apache Spark Bayes compress-zstd: 19, Long Mode - Decompression Speed srsran: 4G PHY_DL_Test 100 PRB SISO 64-QAM renaissance: Rand Forest jpegxl: PNG - 5 botan: KASUMI - Decrypt jpegxl-decode: All srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM vpxenc: Speed 5 - Bosphorus 4K stress-ng: Socket Activity botan: Twofish renaissance: ALS Movie Lens renaissance: Genetic Algorithm Using Jenetics + Futures botan: Twofish - Decrypt build-gcc: Time To Compile botan: KASUMI dav1d: Summer Nature 1080p srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM npb: CG.C botan: ChaCha20Poly1305 - Decrypt botan: ChaCha20Poly1305 libgav1: Summer Nature 1080p srsran: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM embree: Pathtracer ISPC - Asian Dragon srsran: 4G PHY_DL_Test 100 PRB SISO 64-QAM tnn: CPU - SqueezeNet v2 compress-zstd: 8 - Decompression Speed xmrig: Wownero - 1M mnn: mobilenet-v1-1.0 libgav1: Chimera 1080p vpxenc: Speed 0 - Bosphorus 4K ncnn: CPU - blazeface stress-ng: CPU Stress npb: MG.C onnx: fcn-resnet101-11 - OpenMP CPU stress-ng: Forking tjbench: Decompression Throughput build-gdb: Time To Compile npb: SP.C compress-zstd: 3, Long Mode - Compression Speed jpegxl-decode: 1 svt-av1: Preset 4 - Bosphorus 4K jpegxl: JPEG - 5 stress-ng: SENDFILE srsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAM tnn: CPU - DenseNet vpxenc: Speed 0 - Bosphorus 1080p renaissance: Scala Dotty xmrig: Monero - 1M mnn: resnet-v2-50 stress-ng: Matrix Math compress-zstd: 8, Long Mode - Decompression Speed srsran: OFDM_Test compress-zstd: 19, Long Mode - Compression Speed svt-av1: Preset 4 - Bosphorus 1080p svt-av1: Preset 8 - Bosphorus 1080p onnx: super-resolution-10 - OpenMP CPU renaissance: Apache Spark ALS svt-av1: Preset 8 - Bosphorus 4K stress-ng: MEMFD libgav1: Summer Nature 4K tachyon: Total Time tnn: CPU - MobileNet v2 stress-ng: Atomic embree: Pathtracer - Crown jpegxl: PNG - 7 pjsip: OPTIONS, Stateful helsing: 12 digit stress-ng: System V Message Passing draco: Lion srsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAM npb: IS.D dav1d: Chimera 1080p npb: LU.C openssl: RSA4096 stress-ng: Crypto pjsip: OPTIONS, Stateless gromacs: MPI CPU - water_GMX50_bare stress-ng: Semaphores stress-ng: Vector Math libgav1: Chimera 1080p 10-bit compress-zstd: 3, Long Mode - Decompression Speed build-llvm: Ninja openssl: RSA4096 npb: BT.C stress-ng: Malloc dav1d: Chimera 1080p 10-bit onnx: yolov4 - OpenMP CPU pjsip: INVITE npb: FT.C dav1d: Summer Nature 4K ncnn: CPU - vgg16 npb: EP.D npb: EP.C oidn: RTLightmap.hdr.4096x4096 oidn: RT.ldr_alb_nrm.3840x2160 oidn: RT.hdr_alb_nrm.3840x2160 fs-mark: 4000 Files, 32 Sub Dirs, 1MB Size fs-mark: 5000 Files, 1MB Size, 4 Threads fs-mark: 1000 Files, 1MB Size fs-mark: 1000 Files, 1MB Size, No Sync/FSync 1 2 2a 3 4 29.28 108.084 175.761 11161.0 9.32 28.77 9.448 9.51 14.59 11.968 3.4 14.06 4.761 749.5 9162 9.53 21552725790 424.2 29.11 205.3 33.59 101.7 6.727 41.6 107.7 4228.3 375.21 16.2394 17756.2 21.6 61.593 9.86 127.717 154.805 370072.74 25.77 300.54 3056 2453.97 154.626 16.15 43.412 13.6064 13.332 5794.59 5805.567 402 2980.6 0.71 63.97 369.6 509.936 13199.42 9909 309.06 354 236.2 52.114 239.488 469.687 21.51 7108751.78 15.2315 23.3 113.9 246.1 947934.41 692.735 4217.7 470.993 3346.8 1037.6 2937 168.6 796.3 42.83 97.454 244.68 45.6 10.39 8734.81 382.49 7741.6 1749.8 382.844 989.595 100.923 606.85 95.5 9201.87 446.243 448.943 159.35 60 14.5825 334.9 67.4 3437.4 6139.3 4.516 126.33 4.48 3.88 33886.48 18132.48 79 25842.06 179.08094 57.644 10015.41 791.5 42.81 1.296 63.38 258515.31 105.9 2777.534 10.48 1037.1 3649.5 42.292 83838.64 3685.3 83600000 32.9 4.199 53.668 2994 1811.8 14.52 664.52 49.24 56.6045 267.629 206896.6 14.9829 9.49 7200 3.279 8448358.81 6971 324.9 955.73 651.46 44806.58 213661.4 4125.17 40556 1.135 2361373.54 61776.67 49.71 3554.9 475.647 3314.6 37989.42 108806562.44 447.79 254 3695 21463.07 201.27 47.68 657.36 657.91 0.18 0.37 0.37 48.3 92.4 182.3 21.41 137.965 214.781 13624.6 10.86 24.49 9.552 9.47 12.91 12.182 3.5 12.78 4.769 699.5 9067 9.36 21046612100 407.9 30.23 203.3 32.06 102 7.049 41.8 107 4417.7 386.2 16.864 17673.5 22.47 62.383 9.81 132.561 149.259 359224.65 26.71 302.28 3070.9 2368.4 149.274 15.82 44.345 13.1739 13.2093 5957.605 5964.414 393 2897.7 0.72 62.24 364.8 512.036 13530.64 9866 310.56 358.5 241.4 53.102 242.673 477.924 21.08 7010637.99 15.1728 23.2 114.4 241.7 945951.92 678.005 4148.2 477.97 3339.5 1036.5 2996.7 167.6 792.8 42.09 96.459 242.69 45.1 10.47 8743.72 378.63 7684.4 1762.6 378.273 992.731 100.092 608.48 94.4 9182.87 452.627 453.623 160.48 60.2 14.5635 333.9 66.972 3441.2 6195.1 4.578 125.16 4.45 3.83 33451.11 17902.94 78 25788.23 181.082108 57.867 10071.2 783.6 42.78 1.291 63.59 256182.61 107.1 2796.345 10.5 1040.0 3612.2 42.146 83421.14 3671.9 83900000 33 4.197 53.263 2969 1811.4 14.492 667.41 49.64 56.7859 268.434 205506.01 15.0633 9.48 7181 3.282 8474070.68 6970 325.4 949.09 654.71 44806.37 213594.7 4101.26 40467 1.131 2366704.41 61530.72 50 3548.5 477.965 3304.6 38081.62 108565365.53 448.62 253 3691 21480.32 201.12 47.63 657.81 657.4 0.18 0.37 0.37 22.73 137.163 177.476 13058.8 9.07 25.7 9.862 10.81 13.34 12.592 3.3 13.93 4.892 723.2 8968 9.47 20036600240 395.4 30.65 205.7 32.42 103.8 7.117 41.3 109 4218.5 382.69 16.199 18075.9 21.68 64.063 10.2 129.197 153.197 360700.02 25.93 311.42 3160.1 2428.99 153.146 15.78 43.953 13.3927 12.9819 5834.221 5840.449 405 2933.9 0.72 62.24 370.9 498.988 13543.95 10122 309.65 363 237.53 53.397 245.246 466.802 21.02 7105060.11 14.8994 23.13 116.5 240.8 933550.38 684.686 4235.8 468.18 3278.9 1032.1 2949.1 170.9 787.9 42.06 95.713 240.31 45.9 10.38 8593.96 376.053 7815.1 1757.6 376.591 1005.902 99.309 598.76 95.9 9328.47 451.156 454.126 159.17 60.9 14.7794 338.7 67.904 3393.4 6150.7 4.574 126.03 4.48 3.88 33867.83 18046.68 78 25641.32 178.842786 58.362 9955.52 791.4 42.29 1.29 63.06 255601.37 106.9 2808.12 10.48 1034.2 3633.6 42.563 83122.18 3649.7 84100000 32.7 4.163 53.201 2968 1821.2 14.398 667.3 49.57 56.9392 269.635 205451.26 14.9625 9.48 7147 3.303 8481535.86 7020 323.8 949.95 650.17 44581.42 212807.5 4097.75 40534 1.128 2369069.01 61407.59 49.71 3535.8 478.198 3297.3 37895.63 108676457.34 448.33 254 3682 21414.7 200.97 47.72 656.73 657.8 0.18 0.37 0.37 23.34 139.043 176.872 13384.4 9.65 24.32 8.567 9.84 13.8 11.217 3.7 13.14 4.49 689.6 8488 10.09 21253211510 416.5 28.57 193.8 34.02 97.9 6.782 39.7 104 4272.0 370.14 16.6098 17372.0 22.05 61.703 9.92 128.584 153.749 356935.54 26.68 303.06 3049.7 2422.93 153.704 15.63 44.852 13.5573 13.4043 5982.508 5986.606 404 2970.7 0.73 63.14 361 508.786 13419.61 10035 317.04 358.9 242.07 52.327 242.847 468.86 21.07 6946969.59 15.2426 23.66 115.2 243.7 927549.4 684.125 4231.4 469.437 3306.9 1053.2 2994.3 170.7 802.9 42.81 96.342 240.99 45.3 10.29 8655.2 377.661 7782.9 1779.1 378.987 994.516 99.982 608.27 95.2 9257.58 453.191 455.885 158.1 60.7 14.5969 338.8 67.941 3409.5 6109.5 4.537 124.62 4.51 3.83 33646.31 18037.85 79 25520.9 179.683632 58.08 9947.95 793.3 42.61 1.281 62.86 256721.13 106.8 2782.519 10.59 1044.9 3644.9 42.143 83933.21 3670.9 83300000 32.7 4.201 53.313 2985 1826.9 14.441 661.85 49.49 56.49 267.522 205290.52 15.0739 9.42 7164 3.287 8420655.77 6981 326.1 950.99 654.35 44890.33 214236 4113.7 40287 1.134 2354743.34 61760.27 49.85 3535.4 476.66 3312.4 37953.5 108285042.75 449.66 253 3685 21415.64 200.7 47.76 657.65 657.82 0.18 0.37 0.37 OpenBenchmarking.org
NCNN Target: CPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: googlenet 2a 3 4 1 7 14 21 28 35 21.41 22.73 23.34 29.28 MIN: 20.02 / MAX: 79.76 MIN: 20.24 / MAX: 145.71 MIN: 20.06 / MAX: 244.6 MIN: 20.2 / MAX: 324.23 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
YafaRay Total Time For Sample Scene OpenBenchmarking.org Seconds, Fewer Is Better YafaRay 3.5.1 Total Time For Sample Scene 1 3 2a 4 30 60 90 120 150 108.08 137.16 137.97 139.04 1. (CXX) g++ options: -std=c++11 -pthread -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype
Timed MrBayes Analysis Primate Phylogeny Analysis OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis 1 4 3 2a 50 100 150 200 250 175.76 176.87 177.48 214.78 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm
Renaissance Test: Savina Reactors.IO OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Savina Reactors.IO 1 3 4 2a 3K 6K 9K 12K 15K 11161.0 13058.8 13384.4 13624.6 MAX: 18984.63 MAX: 22551.43 MAX: 21982.47 MIN: 13624.57 / MAX: 22746.63
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: mnasnet 3 1 4 2a 3 6 9 12 15 9.07 9.32 9.65 10.86 MIN: 8.61 / MAX: 77.83 MIN: 8.35 / MAX: 124.81 MIN: 8.63 / MAX: 109.11 MIN: 8.21 / MAX: 292.11 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: CPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: regnety_400m 4 2a 3 1 7 14 21 28 35 24.32 24.49 25.70 28.77 MIN: 24.18 / MAX: 26.35 MIN: 24.31 / MAX: 35.97 MIN: 24.34 / MAX: 88.38 MIN: 24.15 / MAX: 413.39 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Mobile Neural Network Model: squeezenetv1.1 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: squeezenetv1.1 4 1 2a 3 3 6 9 12 15 8.567 9.448 9.552 9.862 MIN: 8.38 / MAX: 9.21 MIN: 9.29 / MAX: 9.64 MIN: 9.42 / MAX: 9.74 MIN: 9.72 / MAX: 11.94 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
NCNN Target: CPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU-v3-v3 - Model: mobilenet-v3 2a 1 4 3 3 6 9 12 15 9.47 9.51 9.84 10.81 MIN: 8.63 / MAX: 73.71 MIN: 8.92 / MAX: 73.88 MIN: 8.74 / MAX: 105.4 MIN: 8.62 / MAX: 269.88 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: efficientnet-b0 2a 3 4 1 4 8 12 16 20 12.91 13.34 13.80 14.59 MIN: 12.25 / MAX: 76.55 MIN: 12.62 / MAX: 53.86 MIN: 12.63 / MAX: 84.56 MIN: 12.51 / MAX: 233.03 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Mobile Neural Network Model: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: SqueezeNetV1.0 4 1 2a 3 3 6 9 12 15 11.22 11.97 12.18 12.59 MIN: 11.12 / MAX: 12.82 MIN: 11.87 / MAX: 14.31 MIN: 12.09 / MAX: 12.34 MIN: 12.49 / MAX: 12.75 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Natron Input: Spaceship OpenBenchmarking.org FPS, More Is Better Natron 2.4 Input: Spaceship 4 2a 1 3 0.8325 1.665 2.4975 3.33 4.1625 3.7 3.5 3.4 3.3
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: alexnet 2a 4 3 1 4 8 12 16 20 12.78 13.14 13.93 14.06 MIN: 10.83 / MAX: 59.08 MIN: 11.83 / MAX: 66.46 MIN: 11.95 / MAX: 67.38 MIN: 12.15 / MAX: 65.33 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Mobile Neural Network Model: mobilenetV3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: mobilenetV3 4 1 2a 3 1.1007 2.2014 3.3021 4.4028 5.5035 4.490 4.761 4.769 4.892 MIN: 4.44 / MAX: 4.66 MIN: 4.71 / MAX: 4.92 MIN: 4.71 / MAX: 4.95 MIN: 4.84 / MAX: 5.08 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Zstd Compression Compression Level: 8, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 8, Long Mode - Compression Speed 1 3 2a 4 160 320 480 640 800 749.5 723.2 699.5 689.6 1. (CC) gcc options: -O3 -pthread -lz -llzma
ONNX Runtime Model: shufflenet-v2-10 - Device: OpenMP CPU OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.8.2 Model: shufflenet-v2-10 - Device: OpenMP CPU 1 2a 3 4 2K 4K 6K 8K 10K 9162 9067 8968 8488 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
NCNN Target: CPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU-v2-v2 - Model: mobilenet-v2 2a 3 1 4 3 6 9 12 15 9.36 9.47 9.53 10.09 MIN: 9.15 / MAX: 31.57 MIN: 9.06 / MAX: 73.63 MIN: 9.4 / MAX: 9.85 MIN: 9.96 / MAX: 10.76 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenSSL Algorithm: SHA256 OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.0 Algorithm: SHA256 1 4 2a 3 5000M 10000M 15000M 20000M 25000M 21552725790 21253211510 21046612100 20036600240 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Zstd Compression Compression Level: 8 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 8 - Compression Speed 1 4 2a 3 90 180 270 360 450 424.2 416.5 407.9 395.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
NCNN Target: CPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: resnet50 4 1 2a 3 7 14 21 28 35 28.57 29.11 30.23 30.65 MIN: 28.35 / MAX: 31.07 MIN: 28.42 / MAX: 32.31 MIN: 28.87 / MAX: 118.66 MIN: 29.05 / MAX: 121.45 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM 3 1 2a 4 50 100 150 200 250 205.7 205.3 203.3 193.8 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: yolov4-tiny 2a 3 1 4 8 16 24 32 40 32.06 32.42 33.59 34.02 MIN: 31.27 / MAX: 35.76 MIN: 30.88 / MAX: 35.06 MIN: 31.22 / MAX: 35.03 MIN: 33.28 / MAX: 35.26 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
srsRAN Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM 3 2a 1 4 20 40 60 80 100 103.8 102.0 101.7 97.9 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: MobileNetV2_224 1 4 2a 3 2 4 6 8 10 6.727 6.782 7.049 7.117 MIN: 6.65 / MAX: 6.83 MIN: 6.72 / MAX: 7 MIN: 6.99 / MAX: 7.24 MIN: 7.05 / MAX: 7.28 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Zstd Compression Compression Level: 19 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Compression Speed 2a 1 3 4 10 20 30 40 50 41.8 41.6 41.3 39.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
yquake2 Renderer: Software CPU - Resolution: 1920 x 1080 OpenBenchmarking.org Frames Per Second, More Is Better yquake2 8.0 Renderer: Software CPU - Resolution: 1920 x 1080 3 1 2a 4 20 40 60 80 100 109.0 107.7 107.0 104.0 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
Renaissance Test: Apache Spark PageRank OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Apache Spark PageRank 3 1 4 2a 900 1800 2700 3600 4500 4218.5 4228.3 4272.0 4417.7 MIN: 3837.05 / MAX: 4253.4 MIN: 3893.95 / MAX: 4590.5 MIN: 3928.17 / MAX: 4308 MIN: 3994.16 / MAX: 4579.9
Stress-NG Test: CPU Cache OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: CPU Cache 2a 3 1 4 80 160 240 320 400 386.20 382.69 375.21 370.14 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Asian Dragon 2a 4 1 3 4 8 12 16 20 16.86 16.61 16.24 16.20 MIN: 16.66 / MAX: 17.2 MIN: 16.44 / MAX: 16.97 MIN: 16.03 / MAX: 16.57 MIN: 16 / MAX: 16.57
Renaissance Test: Akka Unbalanced Cobwebbed Tree OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Akka Unbalanced Cobwebbed Tree 4 2a 1 3 4K 8K 12K 16K 20K 17372.0 17673.5 17756.2 18075.9 MIN: 13802.39 MIN: 13923.17 / MAX: 17673.51 MIN: 14154.52 / MAX: 17756.23 MIN: 14486.51
NCNN Target: CPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: mobilenet 1 3 4 2a 5 10 15 20 25 21.60 21.68 22.05 22.47 MIN: 21.18 / MAX: 22.04 MIN: 21.26 / MAX: 22.31 MIN: 21.56 / MAX: 22.96 MIN: 21.89 / MAX: 23.47 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Timed Linux Kernel Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.14 Time To Compile 1 4 2a 3 14 28 42 56 70 61.59 61.70 62.38 64.06
NCNN Target: CPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: shufflenet-v2 2a 1 4 3 3 6 9 12 15 9.81 9.86 9.92 10.20 MIN: 9.72 / MAX: 10.1 MIN: 9.78 / MAX: 9.99 MIN: 9.83 / MAX: 10.07 MIN: 9.94 / MAX: 44.59 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Timed HMMer Search Pfam Database Search OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.2 Pfam Database Search 1 4 3 2a 30 60 90 120 150 127.72 128.58 129.20 132.56 1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm -lmpi
Botan Test: CAST-256 OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: CAST-256 1 4 3 2a 30 60 90 120 150 154.81 153.75 153.20 149.26 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
KeyDB OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.2.0 1 3 2a 4 80K 160K 240K 320K 400K 370072.74 360700.02 359224.65 356935.54 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
NCNN Target: CPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: squeezenet_ssd 1 3 4 2a 6 12 18 24 30 25.77 25.93 26.68 26.71 MIN: 23.83 / MAX: 105.86 MIN: 23.99 / MAX: 110.42 MIN: 24.14 / MAX: 135.76 MIN: 24.2 / MAX: 130.14 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Stress-NG Test: NUMA OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: NUMA 3 4 2a 1 70 140 210 280 350 311.42 303.06 302.28 300.54 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Zstd Compression Compression Level: 3 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 3 - Compression Speed 3 2a 1 4 700 1400 2100 2800 3500 3160.1 3070.9 3056.0 3049.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
Stress-NG Test: Memory Copying OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Memory Copying 1 3 4 2a 500 1000 1500 2000 2500 2453.97 2428.99 2422.93 2368.40 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Botan Test: CAST-256 - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: CAST-256 - Decrypt 1 4 3 2a 30 60 90 120 150 154.63 153.70 153.15 149.27 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
NCNN Target: CPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: resnet18 4 3 2a 1 4 8 12 16 20 15.63 15.78 15.82 16.15 MIN: 15.5 / MAX: 17.54 MIN: 15.64 / MAX: 16.13 MIN: 15.65 / MAX: 17.14 MIN: 15.51 / MAX: 51.1 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Timed FFmpeg Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed FFmpeg Compilation 4.4 Time To Compile 1 3 2a 4 10 20 30 40 50 43.41 43.95 44.35 44.85
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Crown 1 4 3 2a 3 6 9 12 15 13.61 13.56 13.39 13.17 MIN: 13.45 / MAX: 13.88 MIN: 13.41 / MAX: 13.83 MIN: 13.24 / MAX: 13.69 MIN: 13.03 / MAX: 13.48
Embree Binary: Pathtracer ISPC - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon Obj 4 1 2a 3 3 6 9 12 15 13.40 13.33 13.21 12.98 MIN: 13.25 / MAX: 13.62 MIN: 13.19 / MAX: 13.6 MIN: 13.05 / MAX: 13.46 MIN: 12.84 / MAX: 13.27
Botan Test: AES-256 - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: AES-256 - Decrypt 4 2a 3 1 1300 2600 3900 5200 6500 5982.51 5957.61 5834.22 5794.59 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: AES-256 OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: AES-256 4 2a 3 1 1300 2600 3900 5200 6500 5986.61 5964.41 5840.45 5805.57 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
ONNX Runtime Model: bertsquad-10 - Device: OpenMP CPU OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.8.2 Model: bertsquad-10 - Device: OpenMP CPU 3 4 1 2a 90 180 270 360 450 405 404 402 393 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Zstd Compression Compression Level: 19 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Decompression Speed 1 4 3 2a 600 1200 1800 2400 3000 2980.6 2970.7 2933.9 2897.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
JPEG XL libjxl Input: PNG - Encode Speed: 8 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.5 Input: PNG - Encode Speed: 8 4 3 2a 1 0.1643 0.3286 0.4929 0.6572 0.8215 0.73 0.72 0.72 0.71 1. (CXX) g++ options: -funwind-tables -O3 -O2 -lbrotlidec -lbrotlicommon -lbrotlienc -pthread
JPEG XL libjxl Input: JPEG - Encode Speed: 7 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.5 Input: JPEG - Encode Speed: 7 1 4 3 2a 14 28 42 56 70 63.97 63.14 62.24 62.24 1. (CXX) g++ options: -funwind-tables -O3 -O2 -lbrotlidec -lbrotlicommon -lbrotlienc -pthread
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM 3 1 2a 4 80 160 240 320 400 370.9 369.6 364.8 361.0 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
Timed LLVM Compilation Build System: Unix Makefiles OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 12.0 Build System: Unix Makefiles 3 4 1 2a 110 220 330 440 550 498.99 508.79 509.94 512.04
NAS Parallel Benchmarks Test / Class: SP.B OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B 3 2a 4 1 3K 6K 9K 12K 15K 13543.95 13530.64 13419.61 13199.42 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
Google Draco Model: Church Facade OpenBenchmarking.org ms, Fewer Is Better Google Draco 1.4.1 Model: Church Facade 2a 1 4 3 2K 4K 6K 8K 10K 9866 9909 10035 10122 1. (CXX) g++ options: -O3
Stress-NG Test: MMAP OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: MMAP 4 2a 3 1 70 140 210 280 350 317.04 310.56 309.65 309.06 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM 3 4 2a 1 80 160 240 320 400 363.0 358.9 358.5 354.0 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
Stress-NG Test: Glibc Qsort Data Sorting OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Glibc Qsort Data Sorting 4 2a 3 1 50 100 150 200 250 242.07 241.40 237.53 236.20 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: inception-v3 1 4 2a 3 12 24 36 48 60 52.11 52.33 53.10 53.40 MIN: 51.87 / MAX: 60.82 MIN: 52.04 / MAX: 63.79 MIN: 52.83 / MAX: 64.41 MIN: 53.05 / MAX: 65.86 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
TNN Target: CPU - Model: SqueezeNet v1.1 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: SqueezeNet v1.1 1 2a 4 3 50 100 150 200 250 239.49 242.67 242.85 245.25 MIN: 238.57 / MAX: 240.84 MIN: 241.8 / MAX: 243.8 MIN: 241.08 / MAX: 244.2 MIN: 244.14 / MAX: 246.38 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
Botan Test: Blowfish - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Blowfish - Decrypt 2a 1 4 3 100 200 300 400 500 477.92 469.69 468.86 466.80 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
VP9 libvpx Encoding Speed: Speed 5 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.10.0 Speed: Speed 5 - Input: Bosphorus 1080p 1 2a 4 3 5 10 15 20 25 21.51 21.08 21.07 21.02 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
Stress-NG Test: Context Switching OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Context Switching 1 3 2a 4 1.5M 3M 4.5M 6M 7.5M 7108751.78 7105060.11 7010637.99 6946969.59 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Embree Binary: Pathtracer - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Asian Dragon Obj 4 1 2a 3 4 8 12 16 20 15.24 15.23 15.17 14.90 MIN: 15.06 / MAX: 15.56 MIN: 15.03 / MAX: 15.55 MIN: 14.99 / MAX: 15.51 MIN: 14.72 / MAX: 15.19
JPEG XL libjxl Input: JPEG - Encode Speed: 8 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.5 Input: JPEG - Encode Speed: 8 4 1 2a 3 6 12 18 24 30 23.66 23.30 23.20 23.13 1. (CXX) g++ options: -funwind-tables -O3 -O2 -lbrotlidec -lbrotlicommon -lbrotlienc -pthread
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM 3 4 2a 1 30 60 90 120 150 116.5 115.2 114.4 113.9 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
yquake2 Renderer: OpenGL 3.x - Resolution: 1920 x 1080 OpenBenchmarking.org Frames Per Second, More Is Better yquake2 8.0 Renderer: OpenGL 3.x - Resolution: 1920 x 1080 1 4 2a 3 50 100 150 200 250 246.1 243.7 241.7 240.8 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
Stress-NG Test: Glibc C String Functions OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Glibc C String Functions 1 2a 3 4 200K 400K 600K 800K 1000K 947934.41 945951.92 933550.38 927549.40 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Google SynthMark Test: VoiceMark_100 OpenBenchmarking.org Voices, More Is Better Google SynthMark 20201109 Test: VoiceMark_100 1 3 4 2a 150 300 450 600 750 692.74 684.69 684.13 678.01 1. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast
Renaissance Test: In-Memory Database Shootout OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: In-Memory Database Shootout 2a 1 4 3 900 1800 2700 3600 4500 4148.2 4217.7 4231.4 4235.8 MIN: 3826.63 / MAX: 4701.38 MIN: 3874.71 / MAX: 4758.44 MIN: 3870.51 / MAX: 4488.08 MIN: 3929.76 / MAX: 4695.12
Botan Test: Blowfish OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Blowfish 2a 1 4 3 100 200 300 400 500 477.97 470.99 469.44 468.18 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Renaissance Test: Finagle HTTP Requests OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Finagle HTTP Requests 3 4 2a 1 700 1400 2100 2800 3500 3278.9 3306.9 3339.5 3346.8 MIN: 3090.57 MIN: 3103.31 / MAX: 3360.5 MIN: 3125.4 / MAX: 3381.62 MIN: 3151.48 / MAX: 3408.5
Renaissance Test: Apache Spark Bayes OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Apache Spark Bayes 3 2a 1 4 200 400 600 800 1000 1032.1 1036.5 1037.6 1053.2 MIN: 744.76 / MAX: 1068.99 MIN: 749.86 / MAX: 1039.5 MIN: 755.49 / MAX: 1176.07 MIN: 744.78 / MAX: 1147.66
Zstd Compression Compression Level: 19, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Decompression Speed 2a 4 3 1 600 1200 1800 2400 3000 2996.7 2994.3 2949.1 2937.0 1. (CC) gcc options: -O3 -pthread -lz -llzma
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM 3 4 1 2a 40 80 120 160 200 170.9 170.7 168.6 167.6 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
Renaissance Test: Random Forest OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Random Forest 3 2a 1 4 200 400 600 800 1000 787.9 792.8 796.3 802.9 MIN: 735.87 / MAX: 960.26 MIN: 731.14 / MAX: 937.98 MIN: 741.01 / MAX: 980.54 MIN: 747.01 / MAX: 967.34
JPEG XL libjxl Input: PNG - Encode Speed: 5 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.5 Input: PNG - Encode Speed: 5 1 4 2a 3 10 20 30 40 50 42.83 42.81 42.09 42.06 1. (CXX) g++ options: -funwind-tables -O3 -O2 -lbrotlidec -lbrotlicommon -lbrotlienc -pthread
Botan Test: KASUMI - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: KASUMI - Decrypt 1 2a 4 3 20 40 60 80 100 97.45 96.46 96.34 95.71 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
JPEG XL Decoding libjxl CPU Threads: All OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.5 CPU Threads: All 1 2a 4 3 50 100 150 200 250 244.68 242.69 240.99 240.31
srsRAN Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM 3 1 4 2a 10 20 30 40 50 45.9 45.6 45.3 45.1 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
VP9 libvpx Encoding Speed: Speed 5 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.10.0 Speed: Speed 5 - Input: Bosphorus 4K 2a 1 3 4 3 6 9 12 15 10.47 10.39 10.38 10.29 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
Stress-NG Test: Socket Activity OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Socket Activity 2a 1 4 3 2K 4K 6K 8K 10K 8743.72 8734.81 8655.20 8593.96 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Botan Test: Twofish OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Twofish 1 2a 4 3 80 160 240 320 400 382.49 378.63 377.66 376.05 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Renaissance Test: ALS Movie Lens OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: ALS Movie Lens 2a 1 4 3 2K 4K 6K 8K 10K 7684.4 7741.6 7782.9 7815.1 MAX: 8257.35 MIN: 7715.52 / MAX: 8202.29 MIN: 7780.5 / MAX: 8351.41 MIN: 7715.83 / MAX: 8375.58
Renaissance Test: Genetic Algorithm Using Jenetics + Futures OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Genetic Algorithm Using Jenetics + Futures 1 3 2a 4 400 800 1200 1600 2000 1749.8 1757.6 1762.6 1779.1 MIN: 1674.68 / MAX: 1786.55 MIN: 1710.78 / MAX: 1813.38 MIN: 1712.52 / MAX: 1808.91 MIN: 1744.3 / MAX: 1933.19
Botan Test: Twofish - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Twofish - Decrypt 1 4 2a 3 80 160 240 320 400 382.84 378.99 378.27 376.59 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Timed GCC Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GCC Compilation 11.2.0 Time To Compile 1 2a 4 3 200 400 600 800 1000 989.60 992.73 994.52 1005.90
Botan Test: KASUMI OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: KASUMI 1 2a 4 3 20 40 60 80 100 100.92 100.09 99.98 99.31 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
dav1d Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.9.2 Video Input: Summer Nature 1080p 2a 4 1 3 130 260 390 520 650 608.48 608.27 606.85 598.76 MIN: 437.38 / MAX: 657.25 MIN: 441.24 / MAX: 660.61 MIN: 437.62 / MAX: 657.32 MIN: 375.79 / MAX: 650.29 1. (CC) gcc options: -lm -pthread
srsRAN Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM 3 1 4 2a 20 40 60 80 100 95.9 95.5 95.2 94.4 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
NAS Parallel Benchmarks Test / Class: CG.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C 3 4 1 2a 2K 4K 6K 8K 10K 9328.47 9257.58 9201.87 9182.87 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
Botan Test: ChaCha20Poly1305 - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: ChaCha20Poly1305 - Decrypt 4 2a 3 1 100 200 300 400 500 453.19 452.63 451.16 446.24 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: ChaCha20Poly1305 OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: ChaCha20Poly1305 4 3 2a 1 100 200 300 400 500 455.89 454.13 453.62 448.94 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
libgav1 Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better libgav1 0.16.3 Video Input: Summer Nature 1080p 2a 1 3 4 40 80 120 160 200 160.48 159.35 159.17 158.10 1. (CXX) g++ options: -O3 -lpthread -lrt
srsRAN Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 5G PHY_DL_NR Test 270 PRB SISO 256-QAM 3 4 2a 1 14 28 42 56 70 60.9 60.7 60.2 60.0 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon 3 4 1 2a 4 8 12 16 20 14.78 14.60 14.58 14.56 MIN: 14.62 / MAX: 15.1 MIN: 14.43 / MAX: 14.91 MIN: 14.41 / MAX: 14.87 MIN: 14.4 / MAX: 14.83
srsRAN Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM 4 3 1 2a 70 140 210 280 350 338.8 338.7 334.9 333.9 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
TNN Target: CPU - Model: SqueezeNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: SqueezeNet v2 2a 1 3 4 15 30 45 60 75 66.97 67.40 67.90 67.94 MIN: 66.65 / MAX: 67.43 MIN: 66.67 / MAX: 68.03 MIN: 67.46 / MAX: 68.31 MIN: 67.55 / MAX: 68.66 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
Zstd Compression Compression Level: 8 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 8 - Decompression Speed 2a 1 4 3 700 1400 2100 2800 3500 3441.2 3437.4 3409.5 3393.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
Xmrig Variant: Wownero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.12.1 Variant: Wownero - Hash Count: 1M 2a 3 1 4 1300 2600 3900 5200 6500 6195.1 6150.7 6139.3 6109.5 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Mobile Neural Network Model: mobilenet-v1-1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: mobilenet-v1-1.0 1 4 3 2a 1.0301 2.0602 3.0903 4.1204 5.1505 4.516 4.537 4.574 4.578 MIN: 4.45 / MAX: 4.68 MIN: 4.48 / MAX: 6.12 MIN: 4.52 / MAX: 4.72 MIN: 4.53 / MAX: 4.75 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
libgav1 Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better libgav1 0.16.3 Video Input: Chimera 1080p 1 3 2a 4 30 60 90 120 150 126.33 126.03 125.16 124.62 1. (CXX) g++ options: -O3 -lpthread -lrt
VP9 libvpx Encoding Speed: Speed 0 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.10.0 Speed: Speed 0 - Input: Bosphorus 4K 4 3 1 2a 1.0148 2.0296 3.0444 4.0592 5.074 4.51 4.48 4.48 4.45 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: blazeface 2a 4 1 3 0.873 1.746 2.619 3.492 4.365 3.83 3.83 3.88 3.88 MIN: 3.76 / MAX: 3.95 MIN: 3.77 / MAX: 3.99 MIN: 3.84 / MAX: 4.08 MIN: 3.81 / MAX: 4.05 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
Stress-NG Test: CPU Stress OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: CPU Stress 1 3 4 2a 7K 14K 21K 28K 35K 33886.48 33867.83 33646.31 33451.11 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
NAS Parallel Benchmarks Test / Class: MG.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C 1 3 4 2a 4K 8K 12K 16K 20K 18132.48 18046.68 18037.85 17902.94 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
ONNX Runtime Model: fcn-resnet101-11 - Device: OpenMP CPU OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.8.2 Model: fcn-resnet101-11 - Device: OpenMP CPU 4 1 3 2a 20 40 60 80 100 79 79 78 78 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Stress-NG Test: Forking OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Forking 1 2a 3 4 6K 12K 18K 24K 30K 25842.06 25788.23 25641.32 25520.90 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
libjpeg-turbo tjbench Test: Decompression Throughput OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 2.1.0 Test: Decompression Throughput 2a 4 1 3 40 80 120 160 200 181.08 179.68 179.08 178.84 1. (CC) gcc options: -O3 -rdynamic
Timed GDB GNU Debugger Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GDB GNU Debugger Compilation 10.2 Time To Compile 1 2a 4 3 13 26 39 52 65 57.64 57.87 58.08 58.36
NAS Parallel Benchmarks Test / Class: SP.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C 2a 1 3 4 2K 4K 6K 8K 10K 10071.20 10015.41 9955.52 9947.95 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
Zstd Compression Compression Level: 3, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 3, Long Mode - Compression Speed 4 1 3 2a 200 400 600 800 1000 793.3 791.5 791.4 783.6 1. (CC) gcc options: -O3 -pthread -lz -llzma
JPEG XL Decoding libjxl CPU Threads: 1 OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.5 CPU Threads: 1 1 2a 4 3 10 20 30 40 50 42.81 42.78 42.61 42.29
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K 1 2a 3 4 0.2916 0.5832 0.8748 1.1664 1.458 1.296 1.291 1.290 1.281 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
JPEG XL libjxl Input: JPEG - Encode Speed: 5 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.5 Input: JPEG - Encode Speed: 5 2a 1 3 4 14 28 42 56 70 63.59 63.38 63.06 62.86 1. (CXX) g++ options: -funwind-tables -O3 -O2 -lbrotlidec -lbrotlicommon -lbrotlienc -pthread
Stress-NG Test: SENDFILE OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: SENDFILE 1 4 2a 3 60K 120K 180K 240K 300K 258515.31 256721.13 256182.61 255601.37 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM 2a 3 4 1 20 40 60 80 100 107.1 106.9 106.8 105.9 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
TNN Target: CPU - Model: DenseNet OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: DenseNet 1 4 2a 3 600 1200 1800 2400 3000 2777.53 2782.52 2796.35 2808.12 MIN: 2657.56 / MAX: 2875.65 MIN: 2661.61 / MAX: 2881.73 MIN: 2669.17 / MAX: 2912.78 MIN: 2689.71 / MAX: 2908.19 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
VP9 libvpx Encoding Speed: Speed 0 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.10.0 Speed: Speed 0 - Input: Bosphorus 1080p 4 2a 3 1 3 6 9 12 15 10.59 10.50 10.48 10.48 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
Renaissance Test: Scala Dotty OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Scala Dotty 3 1 2a 4 200 400 600 800 1000 1034.2 1037.1 1040.0 1044.9 MIN: 800.4 / MAX: 1396.51 MIN: 783.86 / MAX: 1566.56 MIN: 799.41 / MAX: 1590.9 MIN: 809.26 / MAX: 1645.99
Xmrig Variant: Monero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.12.1 Variant: Monero - Hash Count: 1M 1 4 3 2a 800 1600 2400 3200 4000 3649.5 3644.9 3633.6 3612.2 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Mobile Neural Network Model: resnet-v2-50 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.2 Model: resnet-v2-50 4 2a 1 3 10 20 30 40 50 42.14 42.15 42.29 42.56 MIN: 41.79 / MAX: 54.72 MIN: 41.81 / MAX: 54.24 MIN: 41.97 / MAX: 51.95 MIN: 42.26 / MAX: 53.66 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Stress-NG Test: Matrix Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Matrix Math 4 1 2a 3 20K 40K 60K 80K 100K 83933.21 83838.64 83421.14 83122.18 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Zstd Compression Compression Level: 8, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 8, Long Mode - Decompression Speed 1 2a 4 3 800 1600 2400 3200 4000 3685.3 3671.9 3670.9 3649.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
srsRAN Test: OFDM_Test OpenBenchmarking.org Samples / Second, More Is Better srsRAN 21.04 Test: OFDM_Test 3 2a 1 4 20M 40M 60M 80M 100M 84100000 83900000 83600000 83300000 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
Zstd Compression Compression Level: 19, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Compression Speed 2a 1 4 3 8 16 24 32 40 33.0 32.9 32.7 32.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 4 - Input: Bosphorus 1080p 4 1 2a 3 0.9452 1.8904 2.8356 3.7808 4.726 4.201 4.199 4.197 4.163 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 8 - Input: Bosphorus 1080p 1 4 2a 3 12 24 36 48 60 53.67 53.31 53.26 53.20 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
ONNX Runtime Model: super-resolution-10 - Device: OpenMP CPU OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.8.2 Model: super-resolution-10 - Device: OpenMP CPU 1 4 2a 3 600 1200 1800 2400 3000 2994 2985 2969 2968 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Renaissance Test: Apache Spark ALS OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.12 Test: Apache Spark ALS 2a 1 3 4 400 800 1200 1600 2000 1811.4 1811.8 1821.2 1826.9 MIN: 1706.54 / MAX: 2078.04 MIN: 1604.76 / MAX: 2037.25 MIN: 1705.62 / MAX: 1991.77 MIN: 1605.78 / MAX: 2078.63
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K 1 2a 4 3 4 8 12 16 20 14.52 14.49 14.44 14.40 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
Stress-NG Test: MEMFD OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: MEMFD 2a 3 1 4 140 280 420 560 700 667.41 667.30 664.52 661.85 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
libgav1 Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better libgav1 0.16.3 Video Input: Summer Nature 4K 2a 3 4 1 11 22 33 44 55 49.64 49.57 49.49 49.24 1. (CXX) g++ options: -O3 -lpthread -lrt
Tachyon Total Time OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.99b6 Total Time 4 1 2a 3 13 26 39 52 65 56.49 56.60 56.79 56.94 1. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
TNN Target: CPU - Model: MobileNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.3 Target: CPU - Model: MobileNet v2 4 1 2a 3 60 120 180 240 300 267.52 267.63 268.43 269.64 MIN: 257.82 / MAX: 276.65 MIN: 257.16 / MAX: 274.57 MIN: 261.96 / MAX: 278.76 MIN: 256.59 / MAX: 280.11 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
Stress-NG Test: Atomic OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Atomic 1 2a 3 4 40K 80K 120K 160K 200K 206896.60 205506.01 205451.26 205290.52 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Crown 4 2a 1 3 4 8 12 16 20 15.07 15.06 14.98 14.96 MIN: 14.89 / MAX: 15.43 MIN: 14.88 / MAX: 15.44 MIN: 14.8 / MAX: 15.33 MIN: 14.79 / MAX: 15.28
JPEG XL libjxl Input: PNG - Encode Speed: 7 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.5 Input: PNG - Encode Speed: 7 1 3 2a 4 3 6 9 12 15 9.49 9.48 9.48 9.42 1. (CXX) g++ options: -funwind-tables -O3 -O2 -lbrotlidec -lbrotlicommon -lbrotlienc -pthread
PJSIP Method: OPTIONS, Stateful OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: OPTIONS, Stateful 1 2a 4 3 1500 3000 4500 6000 7500 7200 7181 7164 7147 1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lopus -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
Helsing Digit Range: 12 digit OpenBenchmarking.org Seconds, Fewer Is Better Helsing 1.0-beta Digit Range: 12 digit 1 2a 4 3 0.7432 1.4864 2.2296 2.9728 3.716 3.279 3.282 3.287 3.303 1. (CC) gcc options: -O2 -pthread
Stress-NG Test: System V Message Passing OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: System V Message Passing 3 2a 1 4 2M 4M 6M 8M 10M 8481535.86 8474070.68 8448358.81 8420655.77 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Google Draco Model: Lion OpenBenchmarking.org ms, Fewer Is Better Google Draco 1.4.1 Model: Lion 2a 1 4 3 1500 3000 4500 6000 7500 6970 6971 6981 7020 1. (CXX) g++ options: -O3
srsRAN Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 21.04 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM 4 2a 1 3 70 140 210 280 350 326.1 325.4 324.9 323.8 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lm -lfftw3f
NAS Parallel Benchmarks Test / Class: IS.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D 1 4 3 2a 200 400 600 800 1000 955.73 950.99 949.95 949.09 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
dav1d Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.9.2 Video Input: Chimera 1080p 2a 4 1 3 140 280 420 560 700 654.71 654.35 651.46 650.17 MIN: 509.25 / MAX: 838.07 MIN: 508.07 / MAX: 833.66 MIN: 507.78 / MAX: 817.93 MIN: 507.28 / MAX: 822.69 1. (CC) gcc options: -lm -pthread
NAS Parallel Benchmarks Test / Class: LU.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C 4 1 2a 3 10K 20K 30K 40K 50K 44890.33 44806.58 44806.37 44581.42 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.0 Algorithm: RSA4096 4 1 2a 3 50K 100K 150K 200K 250K 214236.0 213661.4 213594.7 212807.5 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Stress-NG Test: Crypto OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Crypto 1 4 2a 3 900 1800 2700 3600 4500 4125.17 4113.70 4101.26 4097.75 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
PJSIP Method: OPTIONS, Stateless OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: OPTIONS, Stateless 1 3 2a 4 9K 18K 27K 36K 45K 40556 40534 40467 40287 1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lopus -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
GROMACS Implementation: MPI CPU - Input: water_GMX50_bare OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2021.2 Implementation: MPI CPU - Input: water_GMX50_bare 1 4 2a 3 0.2554 0.5108 0.7662 1.0216 1.277 1.135 1.134 1.131 1.128 1. (CXX) g++ options: -O3 -pthread
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Semaphores 3 2a 1 4 500K 1000K 1500K 2000K 2500K 2369069.01 2366704.41 2361373.54 2354743.34 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
Stress-NG Test: Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Vector Math 1 4 2a 3 13K 26K 39K 52K 65K 61776.67 61760.27 61530.72 61407.59 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
libgav1 Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better libgav1 0.16.3 Video Input: Chimera 1080p 10-bit 2a 4 3 1 11 22 33 44 55 50.00 49.85 49.71 49.71 1. (CXX) g++ options: -O3 -lpthread -lrt
Zstd Compression Compression Level: 3, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 3, Long Mode - Decompression Speed 1 2a 3 4 800 1600 2400 3200 4000 3554.9 3548.5 3535.8 3535.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
Timed LLVM Compilation Build System: Ninja OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 12.0 Build System: Ninja 1 4 2a 3 100 200 300 400 500 475.65 476.66 477.97 478.20
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.0 Algorithm: RSA4096 1 4 2a 3 700 1400 2100 2800 3500 3314.6 3312.4 3304.6 3297.3 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
NAS Parallel Benchmarks Test / Class: BT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C 2a 1 4 3 8K 16K 24K 32K 40K 38081.62 37989.42 37953.50 37895.63 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
Stress-NG Test: Malloc OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Malloc 1 3 2a 4 20M 40M 60M 80M 100M 108806562.44 108676457.34 108565365.53 108285042.75 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lsctp -lz -ldl -pthread -lc -latomic
dav1d Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better dav1d 0.9.2 Video Input: Chimera 1080p 10-bit 4 2a 3 1 100 200 300 400 500 449.66 448.62 448.33 447.79 MIN: 349.28 / MAX: 620.9 MIN: 348.69 / MAX: 623.43 MIN: 348.83 / MAX: 616.96 MIN: 348.67 / MAX: 626.63 1. (CC) gcc options: -lm -pthread
ONNX Runtime Model: yolov4 - Device: OpenMP CPU OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.8.2 Model: yolov4 - Device: OpenMP CPU 3 1 4 2a 60 120 180 240 300 254 254 253 253 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
PJSIP Method: INVITE OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: INVITE 1 2a 4 3 800 1600 2400 3200 4000 3695 3691 3685 3682 1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lopus -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
NAS Parallel Benchmarks Test / Class: FT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C 2a 1 4 3 5K 10K 15K 20K 25K 21480.32 21463.07 21415.64 21414.70 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better dav1d 0.9.2 Video Input: Summer Nature 4K 1 2a 3 4 40 80 120 160 200 201.27 201.12 200.97 200.70 MIN: 153.49 / MAX: 211.85 MIN: 153.98 / MAX: 211.78 MIN: 153.51 / MAX: 211.71 MIN: 151.83 / MAX: 211.44 1. (CC) gcc options: -lm -pthread
NCNN Target: CPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20210720 Target: CPU - Model: vgg16 2a 1 3 4 11 22 33 44 55 47.63 47.68 47.72 47.76 MIN: 46.53 / MAX: 61.56 MIN: 46.62 / MAX: 52.32 MIN: 46.63 / MAX: 55.12 MIN: 46.35 / MAX: 76.46 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
NAS Parallel Benchmarks Test / Class: EP.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D 2a 4 1 3 140 280 420 560 700 657.81 657.65 657.36 656.73 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
NAS Parallel Benchmarks Test / Class: EP.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C 1 4 3 2a 140 280 420 560 700 657.91 657.82 657.80 657.40 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
Intel Open Image Denoise Run: RTLightmap.hdr.4096x4096 OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RTLightmap.hdr.4096x4096 4 3 2a 1 0.0405 0.081 0.1215 0.162 0.2025 0.18 0.18 0.18 0.18
Intel Open Image Denoise Run: RT.ldr_alb_nrm.3840x2160 OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.ldr_alb_nrm.3840x2160 4 3 2a 1 0.0833 0.1666 0.2499 0.3332 0.4165 0.37 0.37 0.37 0.37
Intel Open Image Denoise Run: RT.hdr_alb_nrm.3840x2160 OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.hdr_alb_nrm.3840x2160 4 3 2a 1 0.0833 0.1666 0.2499 0.3332 0.4165 0.37 0.37 0.37 0.37
FS-Mark Test: 4000 Files, 32 Sub Dirs, 1MB Size OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 4000 Files, 32 Sub Dirs, 1MB Size 1 11 22 33 44 55 48.3
FS-Mark Test: 5000 Files, 1MB Size, 4 Threads OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 5000 Files, 1MB Size, 4 Threads 1 20 40 60 80 100 92.4
FS-Mark Test: 1000 Files, 1MB Size OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size 1 40 80 120 160 200 182.3
Phoronix Test Suite v10.8.4