Microsoft Azure EPYC 7003 HBv3 Benchmarks

Azure HBv3 vs. Azure HBv2 benchmarks.

HTML result view exported from: https://openbenchmarking.org/result/2104110-IB-2104116PT06&grs.

Microsoft Azure EPYC 7003 HBv3 BenchmarksProcessorMotherboardMemoryDiskGraphicsOSKernelCompilerFile-SystemScreen ResolutionSystem LayerAzure HBv3Azure HBv2Azure HBv12 x AMD EPYC 7V13 64-Core (120 Cores)Microsoft Virtual Machine (Hyper-V UEFI v4.1 BIOS)442GB2 x 960GB Microsoft NVMe Direct Disk + 32GB Virtual Disk + 515GB Virtual Diskhyperv_fbCentOS Linux 84.18.0-147.8.1.el8_1.x86_64 (x86_64)GCC 8.3.1 20190507nfs1152x864microsoft2 x AMD EPYC 7V12 64-Core (120 Cores)Microsoft Virtual Machine (Hyper-V UEFI v4.0 BIOS)450GB960GB Microsoft NVMe Direct Disk + 32GB Virtual Disk + 515GB Virtual Disk2 x AMD EPYC 7551 32-Core (60 Cores)226GB32GB Virtual Disk + 752GB Virtual DiskOpenBenchmarking.orgKernel Details- Transparent Huge Pages: alwaysCompiler Details- --build=x86_64-redhat-linux --disable-libmpx --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=x86-64 --with-gcc-major-version-only --with-isl --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver Processor Details- CPU Microcode: 0xffffffffPython Details- Python 3.6.8Security Details- Azure HBv3: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline STIBP: disabled RSB filling + tsx_async_abort: Not affected - Azure HBv2: 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 generic retpoline STIBP: disabled RSB filling + tsx_async_abort: Not affected - Azure HBv1: 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 generic retpoline STIBP: disabled RSB filling + tsx_async_abort: Not affected

Microsoft Azure EPYC 7003 HBv3 Benchmarksonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUcompress-zstd: 8 - Compression Speedpennant: leblancbiggromacs: Water Benchmarkpennant: sedovbigsvt-vp9: Visual Quality Optimized - Bosphorus 1080ponednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUnamd: ATPase Simulation - 327,506 Atomsonednn: Deconvolution Batch shapes_3d - f32 - CPUrodinia: OpenMP LavaMDsvt-av1: Enc Mode 8 - 1080pnpb: LU.Cmnn: inception-v3botan: ChaCha20Poly1305botan: ChaCha20Poly1305 - Decryptlulesh: plaidml: No - Inference - VGG19 - CPUsvt-hevc: 1 - Bosphorus 1080ponednn: IP Shapes 1D - f32 - CPUsvt-av1: Enc Mode 0 - 1080pplaidml: No - Inference - VGG16 - CPUfinancebench: Bonds OpenMPcompress-zstd: 19 - Decompression Speedcompress-zstd: 19, Long Mode - Decompression Speedquantlib: financebench: Repo OpenMPcompress-zstd: 8, Long Mode - Decompression Speedrodinia: OpenMP HotSpot3Dbuild-nodejs: Time To Compilebuild-llvm: Time To Compilebotan: AES-256botan: AES-256 - Decryptcompress-zstd: 8, Long Mode - Compression Speedgmpbench: Total Timebotan: Twofishbotan: Twofish - Decryptincompact3d: X3D-benchmarking input.i3dbotan: Blowfish - Decryptbotan: Blowfishtnn: CPU - SqueezeNet v1.1mafft: Multiple Sequence Alignment - LSU RNAonednn: Deconvolution Batch shapes_1d - f32 - CPUbotan: CAST-256botan: KASUMIbotan: CAST-256 - Decryptbotan: KASUMI - Decryptbuild-linux-kernel: Time To Compilerodinia: OpenMP Leukocytehpcg: hmmer: Pfam Database Searchncnn: CPU - regnety_400mncnn: CPU - resnet18ncnn: CPU - googlenetcompress-zstd: 8 - Decompression Speedncnn: CPU - squeezenet_ssdncnn: CPU - yolov4-tinyncnn: CPU - blazefacencnn: CPU - mnasnetncnn: CPU-v3-v3 - mobilenet-v3kripke: plaidml: No - Inference - ResNet 50 - CPUncnn: CPU - resnet50ncnn: CPU - alexnetncnn: CPU - vgg16ncnn: CPU - efficientnet-b0ncnn: CPU - shufflenet-v2ncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - mobilenetmnn: mobilenet-v1-1.0mnn: MobileNetV2_224mnn: resnet-v2-50mnn: SqueezeNetV1.0tensorflow-lite: SqueezeNetonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUonednn: IP Shapes 3D - f32 - CPUsvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080psvt-vp9: VMAF Optimized - Bosphorus 1080psvt-hevc: 10 - Bosphorus 1080psvt-hevc: 7 - Bosphorus 1080psvt-av1: Enc Mode 4 - 1080pcompress-zstd: 19, Long Mode - Compression Speedcompress-zstd: 19 - Compression Speedrodinia: OpenMP Streamclusterrodinia: OpenMP CFD Solvercloverleaf: Lagrangian-Eulerian Hydrodynamicsminife: SmallAzure HBv3Azure HBv2Azure HBv10.4556823184.23.3372019.0415.833419337.710.4452230.275661.5818237.80394.02856682.8234.710667.765658.37041636.60334.5345.390.8813070.16138.39104884.2942713717.53727.52299.359196.5468754256.882.489111.327163.7605412.1315407.477771.94893.4346.319345.776287.599508425.158424.738272.61414.3269.50445133.69187.480133.66584.05942.00849.35339.0620175.239823418656.2730.036.8947.6721.0413.7123.2427.873.9504.78229.0868.22466320.4540.194843.1030.255501561.773962.336530.898875.1660.4066860.5752380.3838553.21204376.91357.06548.33378.0112.27536.378.27.3798.46016.6613785.30.7321302653.13.4865038.4586.026449140.590.5247370.300561.5955739.19878.10453829.3455.356615.795610.86134803.53923.1236.121.124840.12425.57133719.3281252631.02674.21725.275154.2578133037.5107.976118.969174.7433934.4013938.444588.04155.5280.902283.257318.905009349.147348.425323.61516.06412.1051113.60073.175113.72471.10144.92257.04037.2686207.867438555505.6064.8010.80133.7832.4221.3937.2052.568.03012.52752.53314.13674885.4796.8161294.530.350347791.4601287.03778.3051314.001.0319420.5819810.4410129.68321163.51140.30379.91166.909.51233.469.512.91213.04223.7813165.092.73621920.110.337643.28815.52899129.441.154650.709473.9407790.17540.14625304.1976.799303.243300.17819011.45815.9721.121.878340.07619.88195702.7968752005.42019.31251.7108606.5026042339.6145.569194.550283.2913144.3133151.228468.83021.3217.312218.034455.812378272.733273.039420.04722.0677.9156988.54557.95088.55555.98058.73867.29630.4314209.386111.0539.2441.492163.246.5465.497.9916.6817.35387690974.8063.2526.88147.8823.1719.2618.5043.126.2249.80262.44617.19495206.23423.155355.241.512713373.885539.483344.145524.502.405618.867521.7635510.22263141.53132.23141.93119.204.66925.148.544.06512.94025.0011999.57OpenBenchmarking.org

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv10.61561.23121.84682.46243.078SE +/- 0.002442, N = 3SE +/- 0.002363, N = 3SE +/- 0.005596, N = 30.4556820.7321302.736210MIN: 0.38MIN: 0.65MIN: 2.71. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

Zstd Compression

Compression Level: 8 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Compression SpeedAzure HBv3Azure HBv2Azure HBv17001400210028003500SE +/- 45.10, N = 3SE +/- 33.03, N = 15SE +/- 6.06, N = 33184.22653.1920.11. (CC) gcc options: -O3 -pthread -lz -llzma

Pennant

Test: leblancbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigAzure HBv3Azure HBv2Azure HBv13691215SE +/- 0.017738, N = 3SE +/- 0.009939, N = 3SE +/- 0.027196, N = 33.3372013.48650310.3376401. (CXX) g++ options: -fopenmp -fexceptions -pthread -lmpi_cxx -lmpi

GROMACS

Water Benchmark

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkAzure HBv3Azure HBv2Azure HBv13691215SE +/- 0.009, N = 3SE +/- 0.009, N = 3SE +/- 0.025, N = 159.0418.4583.2881. (CXX) g++ options: -O2 -pthread -lrt -lpthread -lm

Pennant

Test: sedovbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigAzure HBv3Azure HBv2Azure HBv148121620SE +/- 0.003905, N = 3SE +/- 0.012548, N = 3SE +/- 0.085157, N = 35.8334196.02644915.5289901. (CXX) g++ options: -fopenmp -fexceptions -pthread -lmpi_cxx -lmpi

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080pAzure HBv3Azure HBv2Azure HBv170140210280350SE +/- 4.36, N = 3SE +/- 0.20, N = 3SE +/- 1.10, N = 8337.71140.59129.441. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv10.25980.51960.77941.03921.299SE +/- 0.004912, N = 4SE +/- 0.004967, N = 3SE +/- 0.003421, N = 30.4452230.5247371.154650MIN: 0.39MIN: 0.48MIN: 1.11. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

NAMD

ATPase Simulation - 327,506 Atoms

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsAzure HBv3Azure HBv2Azure HBv10.15960.31920.47880.63840.798SE +/- 0.00027, N = 3SE +/- 0.00059, N = 3SE +/- 0.00507, N = 30.275660.300560.70947

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv10.88671.77342.66013.54684.4335SE +/- 0.00509, N = 3SE +/- 0.01470, N = 3SE +/- 0.00862, N = 31.581821.595573.94077MIN: 1.49MIN: 1.5MIN: 3.871. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDAzure HBv3Azure HBv2Azure HBv120406080100SE +/- 0.23, N = 3SE +/- 0.20, N = 3SE +/- 0.39, N = 337.8039.2090.181. (CXX) g++ options: -O2 -lOpenCL

SVT-AV1

Encoder Mode: Enc Mode 8 - Input: 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080pAzure HBv3Azure HBv2Azure HBv120406080100SE +/- 0.34, N = 3SE +/- 0.65, N = 3SE +/- 0.54, N = 1294.0378.1040.151. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

NAS Parallel Benchmarks

Test / Class: LU.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CAzure HBv3Azure HBv2Azure HBv112K24K36K48K60KSE +/- 428.57, N = 14SE +/- 21.24, N = 3SE +/- 114.01, N = 356682.8253829.3425304.191. (F9X) gfortran options: -O3 -march=native -fexceptions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

Mobile Neural Network

Model: inception-v3

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: inception-v3Azure HBv3Azure HBv2Azure HBv120406080100SE +/- 0.24, N = 15SE +/- 0.95, N = 12SE +/- 0.96, N = 1234.7155.3676.80MIN: 31.09 / MAX: 427.77MIN: 47.56 / MAX: 509.74MIN: 69.85 / MAX: 676.141. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl

Botan

Test: ChaCha20Poly1305

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305Azure HBv3Azure HBv2Azure HBv1140280420560700SE +/- 0.27, N = 3SE +/- 0.90, N = 3SE +/- 1.39, N = 3667.77615.80303.241. (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 - DecryptAzure HBv3Azure HBv2Azure HBv1140280420560700SE +/- 0.06, N = 3SE +/- 0.78, N = 3SE +/- 1.37, N = 3658.37610.86300.181. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

LULESH

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3Azure HBv3Azure HBv2Azure HBv19K18K27K36K45KSE +/- 476.10, N = 3SE +/- 44.77, N = 3SE +/- 81.47, N = 341636.6034803.5419011.461. (CXX) g++ options: -O3 -fopenmp -lm -fexceptions -pthread -lmpi_cxx -lmpi

PlaidML

FP16: No - Mode: Inference - Network: VGG19 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPUAzure HBv3Azure HBv2Azure HBv1816243240SE +/- 0.42, N = 4SE +/- 0.31, N = 15SE +/- 0.21, N = 1234.5323.1215.97

SVT-HEVC

Tuning: 1 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pAzure HBv3Azure HBv2Azure HBv11020304050SE +/- 0.60, N = 3SE +/- 0.13, N = 3SE +/- 0.12, N = 345.3936.1221.121. (CC) gcc options: -fPIE -fPIC -O2 -O3 -pie -rdynamic -lpthread -lrt

oneDNN

Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv10.42260.84521.26781.69042.113SE +/- 0.012358, N = 3SE +/- 0.011740, N = 15SE +/- 0.021017, N = 30.8813071.1248401.878340MIN: 0.76MIN: 1.01MIN: 1.751. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

SVT-AV1

Encoder Mode: Enc Mode 0 - Input: 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 0 - Input: 1080pAzure HBv3Azure HBv2Azure HBv10.03620.07240.10860.14480.181SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.000, N = 30.1610.1240.0761. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUAzure HBv3Azure HBv2Azure HBv1918273645SE +/- 0.39, N = 3SE +/- 0.21, N = 3SE +/- 0.21, N = 1538.3925.5719.88

FinanceBench

Benchmark: Bonds OpenMP

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Bonds OpenMPAzure HBv3Azure HBv2Azure HBv140K80K120K160K200KSE +/- 454.05, N = 3SE +/- 476.22, N = 3SE +/- 1191.05, N = 3104884.29133719.33195702.801. (CXX) g++ options: -O3 -march=native -fopenmp

Zstd Compression

Compression Level: 19 - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression SpeedAzure HBv3Azure HBv2Azure HBv18001600240032004000SE +/- 6.14, N = 15SE +/- 1.17, N = 15SE +/- 0.66, N = 153717.52631.02005.41. (CC) gcc options: -O3 -pthread -lz -llzma

Zstd Compression

Compression Level: 19, Long Mode - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Decompression SpeedAzure HBv3Azure HBv2Azure HBv18001600240032004000SE +/- 12.70, N = 15SE +/- 2.70, N = 15SE +/- 0.32, N = 123727.52674.22019.31. (CC) gcc options: -O3 -pthread -lz -llzma

QuantLib

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21Azure HBv3Azure HBv2Azure HBv15001000150020002500SE +/- 4.80, N = 3SE +/- 0.77, N = 3SE +/- 1.79, N = 32299.31725.21251.71. (CXX) g++ options: -O3 -march=native -O2 -rdynamic -lboost_timer -lboost_system -lboost_chrono

FinanceBench

Benchmark: Repo OpenMP

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Repo OpenMPAzure HBv3Azure HBv2Azure HBv120K40K60K80K100KSE +/- 228.72, N = 3SE +/- 166.77, N = 3SE +/- 324.87, N = 359196.5575154.26108606.501. (CXX) g++ options: -O3 -march=native -fopenmp

Zstd Compression

Compression Level: 8, Long Mode - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Decompression SpeedAzure HBv3Azure HBv2Azure HBv19001800270036004500SE +/- 4.89, N = 7SE +/- 4.57, N = 3SE +/- 1.19, N = 154256.83037.52339.61. (CC) gcc options: -O3 -pthread -lz -llzma

Rodinia

Test: OpenMP HotSpot3D

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP HotSpot3DAzure HBv3Azure HBv2Azure HBv1306090120150SE +/- 1.18, N = 15SE +/- 0.89, N = 3SE +/- 1.35, N = 382.49107.98145.571. (CXX) g++ options: -O2 -lOpenCL

Timed Node.js Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 15.11Time To CompileAzure HBv3Azure HBv2Azure HBv14080120160200SE +/- 1.09, N = 3SE +/- 0.83, N = 3SE +/- 0.16, N = 3111.33118.97194.55

Timed LLVM Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileAzure HBv3Azure HBv2Azure HBv160120180240300SE +/- 1.89, N = 3SE +/- 1.99, N = 3SE +/- 3.41, N = 4163.76174.74283.29

Botan

Test: AES-256

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256Azure HBv3Azure HBv2Azure HBv112002400360048006000SE +/- 3.73, N = 3SE +/- 2.68, N = 3SE +/- 16.79, N = 35412.133934.403144.311. (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 - DecryptAzure HBv3Azure HBv2Azure HBv112002400360048006000SE +/- 7.85, N = 3SE +/- 3.26, N = 3SE +/- 1.16, N = 35407.483938.443151.231. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Zstd Compression

Compression Level: 8, Long Mode - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Compression SpeedAzure HBv3Azure HBv2Azure HBv1170340510680850SE +/- 6.99, N = 7SE +/- 0.81, N = 3SE +/- 3.02, N = 15771.9588.0468.81. (CC) gcc options: -O3 -pthread -lz -llzma

GNU GMP GMPbench

Total Time

OpenBenchmarking.orgGMPbench Score, More Is BetterGNU GMP GMPbench 6.2.1Total TimeAzure HBv3Azure HBv2Azure HBv1100020003000400050004893.44155.53021.31. (CC) gcc options: -O3 -fomit-frame-pointer -lm

Botan

Test: Twofish

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: TwofishAzure HBv3Azure HBv2Azure HBv180160240320400SE +/- 1.10, N = 3SE +/- 0.15, N = 3SE +/- 2.54, N = 4346.32280.90217.311. (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 - DecryptAzure HBv3Azure HBv2Azure HBv180160240320400SE +/- 0.77, N = 3SE +/- 0.22, N = 3SE +/- 2.43, N = 4345.78283.26218.031. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Xcompact3d Incompact3d

Input: X3D-benchmarking input.i3d

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: X3D-benchmarking input.i3dAzure HBv3Azure HBv2Azure HBv1100200300400500SE +/- 0.17, N = 3SE +/- 0.07, N = 3SE +/- 2.96, N = 3287.60318.91455.811. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -fexceptions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

Botan

Test: Blowfish - Decrypt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Blowfish - DecryptAzure HBv3Azure HBv2Azure HBv190180270360450SE +/- 0.42, N = 3SE +/- 0.56, N = 3SE +/- 0.08, N = 3425.16349.15272.731. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: Blowfish

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: BlowfishAzure HBv3Azure HBv2Azure HBv190180270360450SE +/- 0.47, N = 3SE +/- 0.42, N = 3SE +/- 0.10, N = 3424.74348.43273.041. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

TNN

Target: CPU - Model: SqueezeNet v1.1

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1Azure HBv3Azure HBv2Azure HBv190180270360450SE +/- 0.15, N = 3SE +/- 0.19, N = 3SE +/- 0.16, N = 3272.61323.62420.05MIN: 272.1 / MAX: 273.5MIN: 322.04 / MAX: 326.05MIN: 418.9 / MAX: 430.091. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O2 -rdynamic -ldl

Timed MAFFT Alignment

Multiple Sequence Alignment - LSU RNA

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.471Multiple Sequence Alignment - LSU RNAAzure HBv3Azure HBv2Azure HBv1510152025SE +/- 0.10, N = 15SE +/- 0.06, N = 3SE +/- 0.24, N = 314.3316.0622.071. (CC) gcc options: -std=c99 -O3 -lm -lpthread

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv13691215SE +/- 0.09389, N = 3SE +/- 0.11811, N = 6SE +/- 0.06125, N = 39.5044512.105107.91569MIN: 4.19MIN: 5.8MIN: 5.111. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

Botan

Test: CAST-256

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256Azure HBv3Azure HBv2Azure HBv1306090120150SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3133.69113.6088.551. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: KASUMI

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMIAzure HBv3Azure HBv2Azure HBv120406080100SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.00, N = 387.4873.1857.951. (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 - DecryptAzure HBv3Azure HBv2Azure HBv1306090120150SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 2133.67113.7288.561. (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 - DecryptAzure HBv3Azure HBv2Azure HBv120406080100SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 384.0671.1055.981. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Timed Linux Kernel Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileAzure HBv3Azure HBv2Azure HBv11326395265SE +/- 0.58, N = 15SE +/- 0.58, N = 13SE +/- 0.45, N = 1342.0144.9258.74

Rodinia

Test: OpenMP Leukocyte

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LeukocyteAzure HBv3Azure HBv2Azure HBv11530456075SE +/- 0.43, N = 8SE +/- 0.57, N = 3SE +/- 0.11, N = 349.3557.0467.301. (CXX) g++ options: -O2 -lOpenCL

High Performance Conjugate Gradient

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Azure HBv3Azure HBv2Azure HBv1918273645SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.17, N = 339.0637.2730.431. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -fexceptions -pthread -lmpi_cxx -lmpi

Timed HMMer Search

Pfam Database Search

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchAzure HBv3Azure HBv2Azure HBv150100150200250SE +/- 1.00, N = 3SE +/- 1.08, N = 3SE +/- 0.42, N = 3175.24207.87209.391. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

NCNN

Target: CPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400mAzure HBv120406080100SE +/- 1.82, N = 12111.05MIN: 101.38 / MAX: 1386.251. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18Azure HBv1918273645SE +/- 0.61, N = 1239.24MIN: 35.91 / MAX: 302.71. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenetAzure HBv1918273645SE +/- 0.52, N = 1241.49MIN: 35.42 / MAX: 661.961. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

Zstd Compression

Compression Level: 8 - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Decompression SpeedAzure HBv15001000150020002500SE +/- 0.12, N = 32163.21. (CC) gcc options: -O3 -pthread -lz -llzma

NCNN

Target: CPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssdAzure HBv11122334455SE +/- 1.11, N = 1246.54MIN: 40.89 / MAX: 1366.371. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tinyAzure HBv11530456075SE +/- 2.55, N = 1265.49MIN: 48.89 / MAX: 1199.51. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazefaceAzure HBv1246810SE +/- 0.15, N = 127.99MIN: 7.33 / MAX: 317.731. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnetAzure HBv148121620SE +/- 0.31, N = 1216.68MIN: 15.15 / MAX: 507.951. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU-v3-v3 - Model: mobilenet-v3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3Azure HBv148121620SE +/- 0.58, N = 1217.35MIN: 14.97 / MAX: 1189.931. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

Kripke

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.4Azure HBv3Azure HBv2Azure HBv120M40M60M80M100MSE +/- 1839362.01, N = 15SE +/- 812916.11, N = 15SE +/- 1483037.21, N = 158234186543855550387690971. (CXX) g++ options: -O2 -fopenmp

PlaidML

FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUAzure HBv3Azure HBv2Azure HBv1246810SE +/- 0.16, N = 9SE +/- 0.03, N = 3SE +/- 0.02, N = 36.275.604.80

NCNN

Target: CPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50Azure HBv3Azure HBv2Azure HBv11428425670SE +/- 0.90, N = 9SE +/- 2.16, N = 12SE +/- 0.82, N = 1230.0364.8063.25MIN: 25.77 / MAX: 962.03MIN: 44.36 / MAX: 2196.77MIN: 49.71 / MAX: 682.191. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnetAzure HBv3Azure HBv2Azure HBv1612182430SE +/- 0.57, N = 9SE +/- 0.21, N = 12SE +/- 0.38, N = 126.8910.8026.88MIN: 5.85 / MAX: 759.55MIN: 9.2 / MAX: 35.44MIN: 24.39 / MAX: 41.821. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg16Azure HBv3Azure HBv2Azure HBv1306090120150SE +/- 2.09, N = 9SE +/- 6.58, N = 12SE +/- 5.55, N = 1247.67133.78147.88MIN: 40.11 / MAX: 1444.13MIN: 82.15 / MAX: 2364.48MIN: 71.72 / MAX: 1164.251. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b0Azure HBv3Azure HBv2Azure HBv1816243240SE +/- 3.62, N = 9SE +/- 1.94, N = 12SE +/- 0.35, N = 1221.0432.4223.17MIN: 12.75 / MAX: 4928.42MIN: 20.32 / MAX: 786.11MIN: 20.64 / MAX: 707.721. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v2Azure HBv3Azure HBv2Azure HBv1510152025SE +/- 0.83, N = 9SE +/- 1.32, N = 12SE +/- 0.31, N = 1213.7121.3919.26MIN: 9.5 / MAX: 286.31MIN: 14.34 / MAX: 1188.39MIN: 17.23 / MAX: 231.691. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU-v2-v2 - Model: mobilenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v2Azure HBv3Azure HBv2Azure HBv1918273645SE +/- 3.10, N = 9SE +/- 2.17, N = 12SE +/- 0.59, N = 1223.2437.2018.50MIN: 10.66 / MAX: 3825.91MIN: 13.35 / MAX: 4343.96MIN: 15.79 / MAX: 651.781. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenetAzure HBv3Azure HBv2Azure HBv11224364860SE +/- 1.74, N = 9SE +/- 2.25, N = 12SE +/- 0.62, N = 1227.8752.5643.12MIN: 21.41 / MAX: 63.52MIN: 40.97 / MAX: 511.3MIN: 38.84 / MAX: 619.91. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread

Mobile Neural Network

Model: mobilenet-v1-1.0

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: mobilenet-v1-1.0Azure HBv3Azure HBv2Azure HBv1246810SE +/- 0.151, N = 15SE +/- 0.500, N = 12SE +/- 0.169, N = 123.9508.0306.224MIN: 2.55 / MAX: 5.41MIN: 4.5 / MAX: 15.23MIN: 5.42 / MAX: 7.731. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl

Mobile Neural Network

Model: MobileNetV2_224

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: MobileNetV2_224Azure HBv3Azure HBv2Azure HBv13691215SE +/- 0.085, N = 15SE +/- 0.587, N = 12SE +/- 0.342, N = 124.78212.5279.802MIN: 3.71 / MAX: 61.38MIN: 7.24 / MAX: 67.34MIN: 8.22 / MAX: 64.561. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl

Mobile Neural Network

Model: resnet-v2-50

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: resnet-v2-50Azure HBv3Azure HBv2Azure HBv11428425670SE +/- 0.25, N = 15SE +/- 1.18, N = 12SE +/- 1.07, N = 1229.0952.5362.45MIN: 25.79 / MAX: 252.29MIN: 41.27 / MAX: 326.88MIN: 54.15 / MAX: 188.441. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl

Mobile Neural Network

Model: SqueezeNetV1.0

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: SqueezeNetV1.0Azure HBv3Azure HBv2Azure HBv148121620SE +/- 0.113, N = 15SE +/- 0.657, N = 12SE +/- 0.162, N = 128.22414.13617.194MIN: 5.66 / MAX: 54.17MIN: 10.98 / MAX: 78.48MIN: 15.74 / MAX: 54.061. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -O2 -rdynamic -pthread -ldl

TensorFlow Lite

Model: SqueezeNet

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetAzure HBv3Azure HBv2Azure HBv120K40K60K80K100KSE +/- 2854.14, N = 15SE +/- 1151.94, N = 15SE +/- 1151.23, N = 1566320.474885.495206.2

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: CPUAzure HBv3Azure HBv2Azure HBv17001400210028003500SE +/- 8.06, N = 15SE +/- 6.35, N = 15SE +/- 87.86, N = 15540.19796.823423.15MIN: 462.7MIN: 726.18MIN: 2651.391. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -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: CPUAzure HBv3Azure HBv2Azure HBv111002200330044005500SE +/- 14.29, N = 15SE +/- 11.10, N = 15SE +/- 122.64, N = 15843.101294.535355.24MIN: 711.66MIN: 1179.19MIN: 4475.571. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv10.34040.68081.02121.36161.702SE +/- 0.000278, N = 3SE +/- 0.006544, N = 12SE +/- 0.088179, N = 150.2555010.3503471.512710MIN: 0.22MIN: 0.3MIN: 0.571. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv17001400210028003500SE +/- 8.80, N = 15SE +/- 5.73, N = 15SE +/- 110.58, N = 12561.77791.463373.88MIN: 471.83MIN: 718.74MIN: 2418.621. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv112002400360048006000SE +/- 102.39, N = 14SE +/- 13.52, N = 15SE +/- 103.35, N = 12962.341287.035539.48MIN: 722.13MIN: 1149.25MIN: 4475.651. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv17001400210028003500SE +/- 8.62, N = 15SE +/- 8.13, N = 3SE +/- 113.65, N = 12530.90778.313344.14MIN: 458.96MIN: 738.78MIN: 2662.071. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv112002400360048006000SE +/- 25.25, N = 12SE +/- 16.54, N = 15SE +/- 83.42, N = 13875.171314.005524.50MIN: 727.38MIN: 1140.59MIN: 4658.381. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv10.54131.08261.62392.16522.7065SE +/- 0.004961, N = 3SE +/- 0.030825, N = 15SE +/- 0.045752, N = 120.4066861.0319422.405610MIN: 0.35MIN: 0.83MIN: 1.961. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv1246810SE +/- 0.005232, N = 3SE +/- 0.000573, N = 3SE +/- 1.542437, N = 150.5752380.5819818.867520MIN: 2.451. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv10.39680.79361.19041.58721.984SE +/- 0.009558, N = 12SE +/- 0.003995, N = 7SE +/- 0.040927, N = 150.3838550.4410121.763550MIN: 0.32MIN: 0.4MIN: 0.891. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

oneDNN

Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUAzure HBv3Azure HBv2Azure HBv13691215SE +/- 0.02162, N = 3SE +/- 0.28898, N = 15SE +/- 0.43651, N = 153.212049.6832110.22263MIN: 2.58MIN: 5.56MIN: 4.11. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pAzure HBv3Azure HBv2Azure HBv180160240320400SE +/- 5.29, N = 15SE +/- 9.44, N = 12SE +/- 0.73, N = 3376.91163.51141.531. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pAzure HBv3Azure HBv2Azure HBv180160240320400SE +/- 23.14, N = 12SE +/- 5.31, N = 15SE +/- 5.97, N = 12357.06140.30132.231. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pAzure HBv3Azure HBv2Azure HBv1120240360480600SE +/- 7.76, N = 3SE +/- 21.86, N = 15SE +/- 2.68, N = 15548.33379.91141.931. (CC) gcc options: -fPIE -fPIC -O2 -O3 -pie -rdynamic -lpthread -lrt

SVT-HEVC

Tuning: 7 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pAzure HBv3Azure HBv2Azure HBv180160240320400SE +/- 4.22, N = 3SE +/- 6.53, N = 12SE +/- 1.15, N = 3378.01166.90119.201. (CC) gcc options: -fPIE -fPIC -O2 -O3 -pie -rdynamic -lpthread -lrt

SVT-AV1

Encoder Mode: Enc Mode 4 - Input: 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080pAzure HBv3Azure HBv2Azure HBv13691215SE +/- 0.031, N = 3SE +/- 0.218, N = 15SE +/- 0.034, N = 1112.2759.5124.6691. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

Zstd Compression

Compression Level: 19, Long Mode - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Compression SpeedAzure HBv3Azure HBv2Azure HBv1816243240SE +/- 0.34, N = 15SE +/- 0.38, N = 15SE +/- 0.69, N = 1236.333.425.11. (CC) gcc options: -O3 -pthread -lz -llzma

Zstd Compression

Compression Level: 19 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression SpeedAzure HBv3Azure HBv2Azure HBv120406080100SE +/- 0.71, N = 15SE +/- 1.18, N = 15SE +/- 0.83, N = 1578.269.548.51. (CC) gcc options: -O3 -pthread -lz -llzma

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP StreamclusterAzure HBv3Azure HBv2Azure HBv11020304050SE +/- 0.196, N = 15SE +/- 0.453, N = 15SE +/- 3.656, N = 157.37912.91244.0651. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP CFD Solver

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP CFD SolverAzure HBv3Azure HBv2Azure HBv13691215SE +/- 0.801, N = 12SE +/- 0.748, N = 12SE +/- 0.361, N = 128.46013.04212.9401. (CXX) g++ options: -O2 -lOpenCL

CloverLeaf

Lagrangian-Eulerian Hydrodynamics

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsAzure HBv3Azure HBv2Azure HBv1612182430SE +/- 0.83, N = 15SE +/- 0.45, N = 12SE +/- 0.11, N = 316.6623.7825.001. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

miniFE

Problem Size: Small

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallAzure HBv3Azure HBv2Azure HBv13K6K9K12K15KSE +/- 387.39, N = 15SE +/- 608.74, N = 12SE +/- 486.67, N = 1513785.3013165.0911999.571. (CXX) g++ options: -O3 -fopenmp -fexceptions -pthread -lmpi_cxx -lmpi

Phoronix Test Suite v10.8.4