LLVM Clang 13 Bencmarking Intel Xeon Ice Lake

LLVM Clang benchmarks for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2108236-TJ-LLVMCLANG38.

LLVM Clang 13 Bencmarking Intel Xeon Ice LakeProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionClang 11Clang 12Clang 132 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads)Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS)Intel Device 0998504GB7682GB INTEL SSDPF2KX076TZASPEEDVE2282 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFPUbuntu 21.045.14.0-rc1-folio (x86_64) 20210715GNOME Shell 3.38.4X Server 1.20.11Clang 11.0.1-2ubuntu4ext41920x1080Clang 12.0.0-3ubuntu1~21.04.1Clang 13.0.0-++20210820072921+23ba3732246a-1~exp1~20210820174536.53OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseEnvironment Details- CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"Processor Details- Scaling Governor: intel_pstate performance - CPU Microcode: 0xd0002a0Security 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected Python Details- Clang 12, Clang 13: Python 3.9.5

LLVM Clang 13 Bencmarking Intel Xeon Ice Lakeblosc: blosclzquantlib: fftw: Float + SSE - 2D FFT Size 4096compress-zstd: 3 - Compression Speedcompress-zstd: 8 - Compression Speedcompress-zstd: 19 - Compression Speedcompress-zstd: 3, Long Mode - Compression Speedcompress-zstd: 8, Long Mode - Compression Speedcompress-zstd: 19, Long Mode - Compression Speedbotan: KASUMIbotan: KASUMI - Decryptbotan: AES-256botan: AES-256 - Decryptbotan: Twofishbotan: Twofish - Decryptbotan: Blowfishbotan: Blowfish - Decryptbotan: CAST-256botan: CAST-256 - Decryptbotan: ChaCha20Poly1305botan: ChaCha20Poly1305 - Decryptlibraw: Post-Processing Benchmarktscp: AI Chess Performancejohn-the-ripper: Blowfishjohn-the-ripper: MD5graphics-magick: Swirlgraphics-magick: Rotategraphics-magick: Sharpengraphics-magick: Enhancedgraphics-magick: Resizinggraphics-magick: Noise-Gaussiangraphics-magick: HWB Color Spacedav1d: Summer Nature 4Kdav1d: Chimera 1080p 10-bitaom-av1: Speed 6 Realtime - Bosphorus 4Kaom-av1: Speed 8 Realtime - Bosphorus 4Kaom-av1: Speed 9 Realtime - Bosphorus 4Ksvt-av1: Preset 4 - Bosphorus 4Ksvt-av1: Preset 8 - Bosphorus 4Ksvt-hevc: 7 - Bosphorus 1080psvt-hevc: 10 - Bosphorus 1080psvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080psvt-vp9: Visual Quality Optimized - Bosphorus 1080pvpxenc: Speed 0 - Bosphorus 4Kvpxenc: Speed 5 - Bosphorus 4Kcoremark: CoreMark Size 666 - Iterations Per Secondpjsip: INVITEpjsip: OPTIONS, Statefulpjsip: OPTIONS, Statelessc-ray: Total Time - 4K, 16 Rays Per Pixelonednn: IP Shapes 1D - bf16bf16bf16 - CPUonednn: IP Shapes 3D - bf16bf16bf16 - CPUonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUaobench: 2048 x 2048 - Total Timeencode-opus: WAV To Opus Encodetachyon: Total Timesynthmark: VoiceMark_100aircrack-ng: securemark: SecureMark-TLSfinancebench: Repo OpenMPfinancebench: Bonds OpenMPmysqlslap: 2048mysqlslap: 4096pgbench: 100 - 250 - Read Onlypgbench: 100 - 250 - Read Only - Average Latencypgbench: 100 - 250 - Read Writepgbench: 100 - 250 - Read Write - Average Latencysqlite-speedtest: Timed Time - Size 1,000draco: Liondraco: Church Facadencnn: CPU - mobilenetncnn: CPU - shufflenet-v2ncnn: CPU - mnasnetncnn: CPU - efficientnet-b0ncnn: CPU - blazefacencnn: CPU - googlenetncnn: CPU - alexnetncnn: CPU - resnet50ncnn: CPU - yolov4-tinyncnn: CPU - squeezenet_ssdncnn: CPU - regnety_400mtnn: CPU - DenseNettnn: CPU - MobileNet v2tnn: CPU - SqueezeNet v2tnn: CPU - SqueezeNet v1.1nginx: 1nginx: 20nginx: 100nginx: 200nginx: 500nginx: 1000apache: 500apache: 1000Clang 11Clang 12Clang 1323751.92589.5181356433.32729.084.8826.7798.845.574.39573.5615743.3475714.916286.034279.963326.769325.704114.754115.648869.352866.09534.551477410113200888000018607287451104473623852528.63842.9519.3244.0654.444.89557.841344.59616.45460.20362.675.6014.322103884.8925215289101954130514.8012.786421.716402.080893.053153.59463608.604368.24836.1679.42314.2482491.027211799.58325208937221.00390658558.2552087003199672140.260851622.93864.4155827700017.3011.4610.9015.326.0721.599.9722.3626.7919.7236.266912.120773.467105.397657.10074606.79254323.92205660.13196465.21204519.00205224.87122576.13120827.4524249.32606.3184436321.02560.680.4865.6833.645.776.91676.2435735.2465700.326293.730288.611329.534326.798115.327114.829882.165874.91136.6514813881113351027533318647237581104452600852532.51844.7819.3944.8454.474.84957.917340.21604.45456.18365.795.6214.112146447.3800825291101854036415.2242.796581.725582.069183.058213.57952611.926368.44936.4289.90813.9533503.930210157.65125084937760.88541758502.8541676983199815250.255841322.97365.2945870705717.6511.6211.0113.756.9421.2411.0421.4828.3519.2637.114353.389541.92783.860402.25575015.29251352.56206297.92194567.38204058.28205740.39121643.80112287.1724671.52657.3184896697.92775.983.6863.1828.047.077.09476.2485761.3035753.637288.807290.469327.933333.278116.103114.500855.764849.95639.9914918161124071043000018747288771174485600866533.02843.9619.6045.3056.654.96758.982353.72626.33476.72374.915.7115.352185330.0367745271101404187112.5942.789101.717702.071683.281143.61489598.020365.72736.0009.63213.4699513.010212825.67724044738153.40429758281.7864587123219734320.257848542.94864.6355967720217.387.848.7912.664.8617.879.6421.7127.9419.2123.984370.099539.64684.610400.26175270.65254133.55208590.36196023.59204880.02205977.68145589.94112600.02OpenBenchmarking.org

C-Blosc

Compressor: blosclz

OpenBenchmarking.orgMB/s, More Is BetterC-Blosc 2.0Compressor: blosclzClang 11Clang 12Clang 135K10K15K20K25KSE +/- 34.48, N = 3SE +/- 34.61, N = 3SE +/- 33.59, N = 323751.924249.324671.51. (CXX) g++ options: -O3 -march=native

QuantLib

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21Clang 11Clang 12Clang 136001200180024003000SE +/- 5.39, N = 3SE +/- 4.59, N = 3SE +/- 8.72, N = 32589.52606.32657.31. (CXX) g++ options: -O3 -march=native -rdynamic

FFTW

Build: Float + SSE - Size: 2D FFT Size 4096

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096Clang 11Clang 12Clang 134K8K12K16K20KSE +/- 187.54, N = 4SE +/- 34.47, N = 3SE +/- 43.73, N = 31813518443184891. (CC) gcc options: -pthread -O3 -march=native -lm

Zstd Compression

Compression Level: 3 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Compression SpeedClang 11Clang 12Clang 1314002800420056007000SE +/- 78.33, N = 3SE +/- 60.99, N = 3SE +/- 47.01, N = 36433.36321.06697.91. (CC) gcc options: -O3 -march=native -pthread -lz -llzma

Zstd Compression

Compression Level: 8 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Compression SpeedClang 11Clang 12Clang 136001200180024003000SE +/- 32.58, N = 3SE +/- 34.31, N = 15SE +/- 7.16, N = 32729.02560.62775.91. (CC) gcc options: -O3 -march=native -pthread -lz -llzma

Zstd Compression

Compression Level: 19 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression SpeedClang 11Clang 12Clang 1320406080100SE +/- 0.85, N = 15SE +/- 0.45, N = 3SE +/- 0.64, N = 1084.880.483.61. (CC) gcc options: -O3 -march=native -pthread -lz -llzma

Zstd Compression

Compression Level: 3, Long Mode - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3, Long Mode - Compression SpeedClang 11Clang 12Clang 132004006008001000SE +/- 1.30, N = 3SE +/- 2.11, N = 3SE +/- 1.29, N = 3826.7865.6863.11. (CC) gcc options: -O3 -march=native -pthread -lz -llzma

Zstd Compression

Compression Level: 8, Long Mode - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8, Long Mode - Compression SpeedClang 11Clang 12Clang 132004006008001000SE +/- 0.29, N = 3SE +/- 1.68, N = 3SE +/- 5.79, N = 3798.8833.6828.01. (CC) gcc options: -O3 -march=native -pthread -lz -llzma

Zstd Compression

Compression Level: 19, Long Mode - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression SpeedClang 11Clang 12Clang 131122334455SE +/- 0.51, N = 4SE +/- 0.55, N = 15SE +/- 0.53, N = 1545.545.747.01. (CC) gcc options: -O3 -march=native -pthread -lz -llzma

Botan

Test: KASUMI

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMIClang 11Clang 12Clang 1320406080100SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 374.4076.9277.091. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: KASUMI - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMI - DecryptClang 11Clang 12Clang 1320406080100SE +/- 0.02, N = 3SE +/- 0.23, N = 3SE +/- 0.02, N = 373.5676.2476.251. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: AES-256

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256Clang 11Clang 12Clang 1312002400360048006000SE +/- 0.30, N = 3SE +/- 5.15, N = 3SE +/- 0.33, N = 35743.355735.255761.301. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: AES-256 - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256 - DecryptClang 11Clang 12Clang 1312002400360048006000SE +/- 0.51, N = 3SE +/- 0.79, N = 3SE +/- 0.20, N = 35714.925700.335753.641. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: Twofish

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: TwofishClang 11Clang 12Clang 1360120180240300SE +/- 0.21, N = 3SE +/- 0.97, N = 3SE +/- 0.22, N = 3286.03293.73288.811. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: Twofish - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Twofish - DecryptClang 11Clang 12Clang 1360120180240300SE +/- 0.36, N = 3SE +/- 0.24, N = 3SE +/- 0.27, N = 3279.96288.61290.471. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: Blowfish

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: BlowfishClang 11Clang 12Clang 1370140210280350SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3326.77329.53327.931. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: Blowfish - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Blowfish - DecryptClang 11Clang 12Clang 1370140210280350SE +/- 0.09, N = 3SE +/- 3.35, N = 3SE +/- 0.04, N = 3325.70326.80333.281. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: CAST-256

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256Clang 11Clang 12Clang 13306090120150SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3114.75115.33116.101. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: CAST-256 - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256 - DecryptClang 11Clang 12Clang 13306090120150SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3115.65114.83114.501. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: ChaCha20Poly1305

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305Clang 11Clang 12Clang 132004006008001000SE +/- 0.40, N = 3SE +/- 0.87, N = 3SE +/- 4.73, N = 3869.35882.17855.761. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: ChaCha20Poly1305 - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305 - DecryptClang 11Clang 12Clang 132004006008001000SE +/- 0.04, N = 3SE +/- 0.95, N = 3SE +/- 1.46, N = 3866.10874.91849.961. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

LibRaw

Post-Processing Benchmark

OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing BenchmarkClang 11Clang 12Clang 13918273645SE +/- 0.13, N = 3SE +/- 0.23, N = 3SE +/- 0.19, N = 334.5536.6539.991. (CXX) g++ options: -O3 -march=native -fopenmp -ljpeg -lz -lm

TSCP

AI Chess Performance

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess PerformanceClang 11Clang 12Clang 13300K600K900K1200K1500KSE +/- 794.00, N = 5SE +/- 1489.72, N = 5SE +/- 806.80, N = 51477410148138814918161. (CC) gcc options: -O3 -march=native

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.9.0-jumbo-1Test: BlowfishClang 11Clang 12Clang 1320K40K60K80K100KSE +/- 249.93, N = 3SE +/- 138.68, N = 3SE +/- 41.83, N = 31132001113351124071. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2

John The Ripper

Test: MD5

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.9.0-jumbo-1Test: MD5Clang 11Clang 12Clang 132M4M6M8M10MSE +/- 8504.90, N = 3SE +/- 16973.84, N = 3SE +/- 3785.94, N = 3888000010275333104300001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2

GraphicsMagick

Operation: Swirl

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SwirlClang 11Clang 12Clang 13400800120016002000SE +/- 13.09, N = 3SE +/- 8.51, N = 3SE +/- 8.69, N = 31860186418741. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Rotate

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: RotateClang 11Clang 12Clang 13160320480640800SE +/- 2.40, N = 3SE +/- 2.65, N = 3SE +/- 7.80, N = 37287237281. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SharpenClang 11Clang 12Clang 132004006008001000SE +/- 3.18, N = 3SE +/- 0.67, N = 3SE +/- 5.24, N = 37457588771. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Enhanced

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: EnhancedClang 11Clang 12Clang 1330060090012001500SE +/- 0.67, N = 3SE +/- 5.49, N = 3SE +/- 5.78, N = 31104110411741. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Resizing

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: ResizingClang 11Clang 12Clang 13100200300400500SE +/- 14.35, N = 12SE +/- 9.74, N = 15SE +/- 8.56, N = 154734524851. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: Noise-Gaussian

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Noise-GaussianClang 11Clang 12Clang 13130260390520650SE +/- 2.33, N = 3SE +/- 4.10, N = 3SE +/- 3.06, N = 36236006001. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

GraphicsMagick

Operation: HWB Color Space

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: HWB Color SpaceClang 11Clang 12Clang 132004006008001000SE +/- 6.84, N = 3SE +/- 5.60, N = 15SE +/- 8.76, N = 38528528661. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

dav1d

Video Input: Summer Nature 4K

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.1Video Input: Summer Nature 4KClang 11Clang 12Clang 13120240360480600SE +/- 2.65, N = 3SE +/- 0.77, N = 3SE +/- 0.44, N = 3528.63532.51533.02MIN: 176.65 / MAX: 587.68MIN: 186.64 / MAX: 587.12MIN: 186.74 / MAX: 587.21. (CC) gcc options: -O3 -march=native -pthread

dav1d

Video Input: Chimera 1080p 10-bit

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.9.1Video Input: Chimera 1080p 10-bitClang 11Clang 12Clang 132004006008001000SE +/- 0.82, N = 3SE +/- 1.52, N = 3SE +/- 2.33, N = 3842.95844.78843.96MIN: 515.35 / MAX: 1115.74MIN: 517.09 / MAX: 1121.33MIN: 503.8 / MAX: 1131.861. (CC) gcc options: -O3 -march=native -pthread

AOM AV1

Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KClang 11Clang 12Clang 13510152025SE +/- 0.23, N = 3SE +/- 0.28, N = 3SE +/- 0.24, N = 319.3219.3919.601. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

AOM AV1

Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KClang 11Clang 12Clang 131020304050SE +/- 0.36, N = 9SE +/- 0.40, N = 3SE +/- 0.42, N = 744.0644.8445.301. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

AOM AV1

Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.1Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KClang 11Clang 12Clang 131326395265SE +/- 0.16, N = 3SE +/- 0.12, N = 3SE +/- 0.03, N = 354.4454.4756.651. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

SVT-AV1

Encoder Mode: Preset 4 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 4KClang 11Clang 12Clang 131.11762.23523.35284.47045.588SE +/- 0.020, N = 3SE +/- 0.004, N = 3SE +/- 0.027, N = 34.8954.8494.9671. (CXX) g++ options: -O3 -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4KClang 11Clang 12Clang 131326395265SE +/- 0.15, N = 3SE +/- 0.14, N = 3SE +/- 0.38, N = 357.8457.9258.981. (CXX) g++ options: -O3 -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

SVT-HEVC

Tuning: 7 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pClang 11Clang 12Clang 1380160240320400SE +/- 2.06, N = 3SE +/- 1.14, N = 3SE +/- 3.85, N = 3344.59340.21353.721. (CC) gcc options: -O3 -march=native -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pClang 11Clang 12Clang 13140280420560700SE +/- 2.02, N = 3SE +/- 2.53, N = 3SE +/- 3.02, N = 3616.45604.45626.331. (CC) gcc options: -O3 -march=native -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pClang 11Clang 12Clang 13100200300400500SE +/- 4.56, N = 3SE +/- 4.57, N = 3SE +/- 5.26, N = 3460.20456.18476.721. (CC) gcc options: -O3 -fcommon -march=native -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080pClang 11Clang 12Clang 1380160240320400SE +/- 1.62, N = 3SE +/- 2.53, N = 3SE +/- 4.27, N = 3362.67365.79374.911. (CC) gcc options: -O3 -fcommon -march=native -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

VP9 libvpx Encoding

Speed: Speed 0 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 4KClang 11Clang 12Clang 131.28482.56963.85445.13926.424SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 35.605.625.711. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=native -fPIC -U_FORTIFY_SOURCE -std=gnu++11

VP9 libvpx Encoding

Speed: Speed 5 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4KClang 11Clang 12Clang 1348121620SE +/- 0.13, N = 3SE +/- 0.09, N = 3SE +/- 0.14, N = 1514.3214.1115.351. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=native -fPIC -U_FORTIFY_SOURCE -std=gnu++11

Coremark

CoreMark Size 666 - Iterations Per Second

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondClang 11Clang 12Clang 13500K1000K1500K2000K2500KSE +/- 2482.75, N = 3SE +/- 3711.92, N = 3SE +/- 1510.09, N = 32103884.892146447.382185330.041. (CC) gcc options: -O2 -O3 -march=native -lrt" -lrt

PJSIP

Method: INVITE

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITEClang 11Clang 12Clang 1311002200330044005500SE +/- 8.50, N = 3SE +/- 10.68, N = 3SE +/- 11.10, N = 35289529152711. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O3 -march=native

PJSIP

Method: OPTIONS, Stateful

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatefulClang 11Clang 12Clang 132K4K6K8K10KSE +/- 4.48, N = 3SE +/- 11.92, N = 3SE +/- 43.11, N = 31019510185101401. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O3 -march=native

PJSIP

Method: OPTIONS, Stateless

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatelessClang 11Clang 12Clang 139K18K27K36K45KSE +/- 575.37, N = 3SE +/- 510.13, N = 3SE +/- 380.52, N = 34130540364418711. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O3 -march=native

C-Ray

Total Time - 4K, 16 Rays Per Pixel

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per PixelClang 11Clang 12Clang 1348121620SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 314.8015.2212.591. (CC) gcc options: -lm -lpthread -O3 -march=native

oneDNN

Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPUClang 11Clang 12Clang 130.62921.25841.88762.51683.146SE +/- 0.00451, N = 3SE +/- 0.00591, N = 3SE +/- 0.00435, N = 32.786422.796582.78910MIN: 2.65MIN: 2.65MIN: 2.651. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPUClang 11Clang 12Clang 130.38830.77661.16491.55321.9415SE +/- 0.00619, N = 3SE +/- 0.00427, N = 3SE +/- 0.00225, N = 31.716401.725581.71770MIN: 1.53MIN: 1.57MIN: 1.551. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUClang 11Clang 12Clang 130.46820.93641.40461.87282.341SE +/- 0.00072, N = 3SE +/- 0.00197, N = 3SE +/- 0.00196, N = 32.080892.069182.07168MIN: 1.99MIN: 1.99MIN: 1.991. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPUClang 11Clang 12Clang 130.73831.47662.21492.95323.6915SE +/- 0.00886, N = 3SE +/- 0.00384, N = 3SE +/- 0.00367, N = 33.053153.058213.28114MIN: 2.87MIN: 2.86MIN: 3.081. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPUClang 11Clang 12Clang 130.81341.62682.44023.25364.067SE +/- 0.00881, N = 3SE +/- 0.00151, N = 3SE +/- 0.00211, N = 33.594633.579523.61489MIN: 3.5MIN: 3.5MIN: 3.521. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUClang 11Clang 12Clang 13130260390520650SE +/- 5.71, N = 13SE +/- 6.83, N = 4SE +/- 0.92, N = 3608.60611.93598.02MIN: 570.79MIN: 569.44MIN: 575.431. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUClang 11Clang 12Clang 1380160240320400SE +/- 2.59, N = 3SE +/- 1.79, N = 3SE +/- 0.40, N = 3368.25368.45365.73MIN: 349.99MIN: 347.27MIN: 355.41. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp=libomp -msse4.1 -fPIC -pie -lpthread -ldl

AOBench

Size: 2048 x 2048 - Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total TimeClang 11Clang 12Clang 13816243240SE +/- 0.03, N = 3SE +/- 0.24, N = 3SE +/- 0.02, N = 336.1736.4336.001. (CC) gcc options: -lm -O3 -march=native

Opus Codec Encoding

WAV To Opus Encode

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodeClang 11Clang 12Clang 133691215SE +/- 0.013, N = 5SE +/- 0.006, N = 5SE +/- 0.069, N = 59.4239.9089.6321. (CXX) g++ options: -O3 -march=native -logg -lm

Tachyon

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.99b6Total TimeClang 11Clang 12Clang 1348121620SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 314.2513.9513.471. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

Google SynthMark

Test: VoiceMark_100

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100Clang 11Clang 12Clang 13110220330440550SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3491.03503.93513.011. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

Aircrack-ng

OpenBenchmarking.orgk/s, More Is BetterAircrack-ng 1.5.2Clang 11Clang 12Clang 1350K100K150K200K250KSE +/- 607.21, N = 3SE +/- 504.25, N = 3SE +/- 467.06, N = 3211799.58210157.65212825.681. (CXX) g++ options: -O3 -fvisibility=hidden -masm=intel -fcommon -rdynamic -lsqlite3 -lpthread -lz -lcrypto -lhwloc -ldl -lm -pthread

SecureMark

Benchmark: SecureMark-TLS

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSClang 11Clang 12Clang 1350K100K150K200K250KSE +/- 87.48, N = 3SE +/- 339.82, N = 3SE +/- 40.95, N = 32520892508492404471. (CC) gcc options: -pedantic -O3

FinanceBench

Benchmark: Repo OpenMP

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Repo OpenMPClang 11Clang 12Clang 138K16K24K32K40KSE +/- 19.15, N = 3SE +/- 290.04, N = 3SE +/- 450.24, N = 437221.0037760.8938153.401. (CXX) g++ options: -O3 -march=native -fopenmp

FinanceBench

Benchmark: Bonds OpenMP

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Bonds OpenMPClang 11Clang 12Clang 1313K26K39K52K65KSE +/- 38.93, N = 3SE +/- 82.28, N = 3SE +/- 17.71, N = 358558.2658502.8558281.791. (CXX) g++ options: -O3 -march=native -fopenmp

MariaDB

Clients: 2048

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.6.4Clients: 2048Clang 11Clang 12Clang 13150300450600750SE +/- 7.30, N = 9SE +/- 6.96, N = 9SE +/- 1.71, N = 3700698712-lbz2 -lsnappy-lpthread -lm -lstdc++-lpthread -lm -lstdc++1. (CXX) g++ options: -fPIC -O3 -march=native -fstack-protector -shared -pthread -ldl -lz -lrt

MariaDB

Clients: 4096

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.6.4Clients: 4096Clang 11Clang 12Clang 1370140210280350SE +/- 1.96, N = 3SE +/- 1.07, N = 3SE +/- 1.31, N = 3319319321-lbz2 -lsnappy-lpthread -lm -lstdc++-lpthread -lm -lstdc++1. (CXX) g++ options: -fPIC -O3 -march=native -fstack-protector -shared -pthread -ldl -lz -lrt

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 250 - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read OnlyClang 11Clang 12Clang 13200K400K600K800K1000KSE +/- 16009.81, N = 15SE +/- 3346.94, N = 3SE +/- 10672.94, N = 39672149815259734321. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average LatencyClang 11Clang 12Clang 130.05850.1170.17550.2340.2925SE +/- 0.005, N = 15SE +/- 0.001, N = 3SE +/- 0.003, N = 30.2600.2550.2571. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 250 - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read WriteClang 11Clang 12Clang 1320K40K60K80K100KSE +/- 786.14, N = 3SE +/- 85.43, N = 3SE +/- 119.61, N = 38516284132848541. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average LatencyClang 11Clang 12Clang 130.66891.33782.00672.67563.3445SE +/- 0.027, N = 3SE +/- 0.003, N = 3SE +/- 0.004, N = 32.9382.9732.9481. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

SQLite Speedtest

Timed Time - Size 1,000

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Clang 11Clang 12Clang 131530456075SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 364.4265.2964.641. (CC) gcc options: -O3 -march=native -ldl -lz -lpthread

Google Draco

Model: Lion

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.4.1Model: LionClang 11Clang 12Clang 1313002600390052006500SE +/- 1.73, N = 35827587059671. (CXX) g++ options: -O3 -march=native

Google Draco

Model: Church Facade

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.4.1Model: Church FacadeClang 11Clang 12Clang 1315003000450060007500SE +/- 3.18, N = 3SE +/- 4.26, N = 37000705772021. (CXX) g++ options: -O3 -march=native

NCNN

Target: CPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mobilenetClang 11Clang 12Clang 1348121620SE +/- 0.27, N = 14SE +/- 0.16, N = 3SE +/- 0.11, N = 317.3017.6517.38MIN: 15.79 / MAX: 37.29MIN: 16.51 / MAX: 19.27MIN: 16.17 / MAX: 40.051. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: shufflenet-v2Clang 11Clang 12Clang 133691215SE +/- 0.67, N = 14SE +/- 1.05, N = 3SE +/- 0.01, N = 311.4611.627.84MIN: 7.56 / MAX: 31.76MIN: 7.71 / MAX: 25.56MIN: 7.49 / MAX: 13.761. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: mnasnetClang 11Clang 12Clang 133691215SE +/- 0.66, N = 14SE +/- 1.31, N = 3SE +/- 1.33, N = 310.9011.018.79MIN: 6.9 / MAX: 32.84MIN: 7.09 / MAX: 20.3MIN: 6.69 / MAX: 26.581. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: efficientnet-b0Clang 11Clang 12Clang 1348121620SE +/- 1.01, N = 14SE +/- 2.00, N = 3SE +/- 2.13, N = 315.3213.7512.66MIN: 8.59 / MAX: 45.64MIN: 9.12 / MAX: 41.76MIN: 8.73 / MAX: 27.071. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: blazefaceClang 11Clang 12Clang 13246810SE +/- 0.47, N = 14SE +/- 1.08, N = 3SE +/- 0.01, N = 36.076.944.86MIN: 4.35 / MAX: 27.78MIN: 4.49 / MAX: 19.88MIN: 4.57 / MAX: 9.451. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: googlenetClang 11Clang 12Clang 13510152025SE +/- 1.02, N = 14SE +/- 1.87, N = 3SE +/- 0.31, N = 321.5921.2417.87MIN: 16.19 / MAX: 50.35MIN: 17.28 / MAX: 39.77MIN: 16.58 / MAX: 37.881. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: alexnetClang 11Clang 12Clang 133691215SE +/- 0.16, N = 14SE +/- 1.31, N = 3SE +/- 0.17, N = 39.9711.049.64MIN: 8.85 / MAX: 29.23MIN: 9.24 / MAX: 14.73MIN: 8.9 / MAX: 11.051. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: resnet50Clang 11Clang 12Clang 13510152025SE +/- 0.61, N = 14SE +/- 0.14, N = 3SE +/- 0.29, N = 322.3621.4821.71MIN: 18.98 / MAX: 71.34MIN: 19.9 / MAX: 40.42MIN: 19.54 / MAX: 44.521. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: yolov4-tinyClang 11Clang 12Clang 13714212835SE +/- 0.26, N = 14SE +/- 0.55, N = 3SE +/- 0.54, N = 326.7928.3527.94MIN: 23.92 / MAX: 56.1MIN: 25.63 / MAX: 60.64MIN: 25.71 / MAX: 39.581. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: squeezenet_ssdClang 11Clang 12Clang 13510152025SE +/- 0.53, N = 14SE +/- 0.22, N = 3SE +/- 0.23, N = 319.7219.2619.21MIN: 17.67 / MAX: 58.39MIN: 18.05 / MAX: 35.83MIN: 18.04 / MAX: 38.361. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

NCNN

Target: CPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: CPU - Model: regnety_400mClang 11Clang 12Clang 13918273645SE +/- 2.48, N = 14SE +/- 3.84, N = 3SE +/- 1.77, N = 336.2637.1123.98MIN: 18.78 / MAX: 124.57MIN: 19.88 / MAX: 96.76MIN: 20.12 / MAX: 66.471. (CXX) g++ options: -O3 -march=native -rdynamic -lomp -lpthread -pthread

TNN

Target: CPU - Model: DenseNet

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNetClang 11Clang 12Clang 1315003000450060007500SE +/- 3.50, N = 3SE +/- 1.66, N = 3SE +/- 2.30, N = 36912.124353.394370.10MIN: 6888.33 / MAX: 6947.75MIN: 4332.87 / MAX: 4442MIN: 4351.72 / MAX: 4450.961. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl

TNN

Target: CPU - Model: MobileNet v2

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2Clang 11Clang 12Clang 13170340510680850SE +/- 0.69, N = 3SE +/- 0.70, N = 3SE +/- 0.66, N = 3773.47541.93539.65MIN: 690.55 / MAX: 820.36MIN: 536.56 / MAX: 554.17MIN: 536.06 / MAX: 559.331. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl

TNN

Target: CPU - Model: SqueezeNet v2

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2Clang 11Clang 12Clang 1320406080100SE +/- 0.15, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 3105.4083.8684.61MIN: 104.83 / MAX: 106.1MIN: 83.28 / MAX: 84.63MIN: 84.23 / MAX: 851. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl

TNN

Target: CPU - Model: SqueezeNet v1.1

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1Clang 11Clang 12Clang 13140280420560700SE +/- 0.05, N = 3SE +/- 0.19, N = 3SE +/- 0.27, N = 3657.10402.26400.26MIN: 656.47 / MAX: 657.89MIN: 401.13 / MAX: 403.41MIN: 399.7 / MAX: 401.331. (CXX) g++ options: -O3 -march=native -fopenmp=libomp -pthread -fvisibility=hidden -fvisibility=default -rdynamic -ldl

nginx

Concurrent Requests: 1

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 1Clang 11Clang 12Clang 1316K32K48K64K80KSE +/- 131.50, N = 3SE +/- 102.54, N = 3SE +/- 160.67, N = 374606.7975015.2975270.651. (CC) gcc options: -ldl -lpthread -lcrypt -lz -O3 -march=native

nginx

Concurrent Requests: 20

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 20Clang 11Clang 12Clang 1350K100K150K200K250KSE +/- 2398.86, N = 3SE +/- 1222.93, N = 3SE +/- 1379.02, N = 3254323.92251352.56254133.551. (CC) gcc options: -ldl -lpthread -lcrypt -lz -O3 -march=native

nginx

Concurrent Requests: 100

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 100Clang 11Clang 12Clang 1340K80K120K160K200KSE +/- 1506.43, N = 3SE +/- 431.14, N = 3SE +/- 987.48, N = 3205660.13206297.92208590.361. (CC) gcc options: -ldl -lpthread -lcrypt -lz -O3 -march=native

nginx

Concurrent Requests: 200

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 200Clang 11Clang 12Clang 1340K80K120K160K200KSE +/- 2823.39, N = 3SE +/- 312.30, N = 3SE +/- 911.83, N = 3196465.21194567.38196023.591. (CC) gcc options: -ldl -lpthread -lcrypt -lz -O3 -march=native

nginx

Concurrent Requests: 500

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 500Clang 11Clang 12Clang 1340K80K120K160K200KSE +/- 554.68, N = 3SE +/- 63.72, N = 3SE +/- 936.49, N = 3204519.00204058.28204880.021. (CC) gcc options: -ldl -lpthread -lcrypt -lz -O3 -march=native

nginx

Concurrent Requests: 1000

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 1000Clang 11Clang 12Clang 1340K80K120K160K200KSE +/- 947.34, N = 3SE +/- 632.41, N = 3SE +/- 1305.06, N = 3205224.87205740.39205977.681. (CC) gcc options: -ldl -lpthread -lcrypt -lz -O3 -march=native

Apache HTTP Server

Concurrent Requests: 500

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 500Clang 11Clang 12Clang 1330K60K90K120K150KSE +/- 798.53, N = 3SE +/- 1145.12, N = 15SE +/- 843.46, N = 3122576.13121643.80145589.941. (CC) gcc options: -shared -fPIC -pthread -O3 -march=native

Apache HTTP Server

Concurrent Requests: 1000

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 1000Clang 11Clang 12Clang 1330K60K90K120K150KSE +/- 924.52, N = 15SE +/- 534.15, N = 3SE +/- 485.15, N = 3120827.45112287.17112600.021. (CC) gcc options: -shared -fPIC -pthread -O3 -march=native

Phoronix Test Suite v10.8.4