Microsoft Azure EPYC 7003 HPC HBv3 Benchmarks

Azure HBv3 benchmarks against other Microsoft Azure HPC instance types. Benchmarks by Michael Larabel for a future article on phoronix.

HTML result view exported from: https://openbenchmarking.org/result/2104153-PTS-AZUREHPC42&sor&grs.

Microsoft Azure EPYC 7003 HPC HBv3 BenchmarksProcessorMotherboardMemoryDiskGraphicsChipsetOSKernelCompilerFile-SystemScreen ResolutionSystem LayerAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure HC - Xeon 8168Azure H - Xeon E5-2667 v32 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 Disk2 x Intel Xeon Platinum 8168 (44 Cores)348GB2 x Intel Xeon E5-2667 v3 (16 Cores)Microsoft Virtual Machine v7.0 (090007 BIOS)Intel 440BX/ZX/DX222GB32GB Virtual Disk + 2199GB Virtual DiskMicrosoft Hyper-V virtual VGAMicrosoft Hyper-V ServerOpenBenchmarking.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 - EPYC 7V13: 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 - EPYC 7V12: 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 - EPYC 7551: 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 HC - Xeon 8168: itlb_multihit: vulnerable + l1tf: Mitigation of PTE Inversion + mds: Mitigation of Clear buffers; SMT Host state unknown + meltdown: Mitigation of PTI + 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: Mitigation of Clear buffers; SMT Host state unknown - Azure H - Xeon E5-2667 v3: itlb_multihit: vulnerable + l1tf: Mitigation of PTE Inversion + mds: Mitigation of Clear buffers; SMT Host state unknown + meltdown: Mitigation of PTI + 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

Microsoft Azure EPYC 7003 HPC HBv3 Benchmarksrodinia: OpenMP LavaMDgromacs: Water Benchmarknpb: LU.Clulesh: svt-hevc: 1 - Bosphorus 1080pcompress-zstd: 8 - Compression Speedbuild-nodejs: Time To Compilesvt-vp9: Visual Quality Optimized - Bosphorus 1080pincompact3d: X3D-benchmarking input.i3dhpcg: build-llvm: Time To Compilesvt-av1: Enc Mode 8 - 1080pbuild-linux-kernel: Time To Compileplaidml: No - Inference - VGG19 - CPUplaidml: No - Inference - VGG16 - CPUbotan: ChaCha20Poly1305botan: ChaCha20Poly1305 - Decryptsvt-av1: Enc Mode 0 - 1080ppennant: leblancbigpennant: sedovbigcompress-zstd: 19 - Decompression Speedcompress-zstd: 19, Long Mode - Decompression Speedbotan: AES-256botan: AES-256 - Decryptcompress-zstd: 8, Long Mode - Decompression Speedfinancebench: Bonds OpenMPquantlib: financebench: Repo OpenMProdinia: OpenMP HotSpot3Dcompress-zstd: 8, Long Mode - Compression Speedgmpbench: Total Timebotan: Twofishbotan: Twofish - Decryptbotan: Blowfish - Decryptbotan: Blowfishtnn: CPU - SqueezeNet v1.1botan: CAST-256botan: KASUMIbotan: CAST-256 - Decryptbotan: KASUMI - Decryptplaidml: No - Inference - ResNet 50 - CPUmnn: inception-v3mnn: resnet-v2-50mnn: SqueezeNetV1.0tensorflow-lite: SqueezeNetonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: 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_3d - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUonednn: IP Shapes 3D - f32 - CPUsvt-hevc: 10 - Bosphorus 1080psvt-hevc: 7 - Bosphorus 1080psvt-av1: Enc Mode 4 - 1080pcompress-zstd: 19 - Compression Speednamd: ATPase Simulation - 327,506 Atomscloverleaf: Lagrangian-Eulerian HydrodynamicsAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure HC - Xeon 8168Azure H - Xeon E5-2667 v337.8039.04156682.8241636.60345.393184.2111.327337.71287.59950839.0620163.76094.02842.00834.5338.39667.765658.3700.1613.3372015.8334193717.53727.55412.1315407.4774256.8104884.2942712299.359196.54687582.489771.94893.4346.319345.776425.158424.738272.614133.69187.480133.66584.0596.2734.71029.0868.22466320.40.445223540.194843.1030.255501561.773962.336530.898875.1660.4556820.4066860.3838553.21204548.33378.0112.27578.20.2756616.6639.1988.45853829.3434803.53936.122653.1118.969140.59318.90500937.2686174.74378.10444.92223.1225.57615.795610.8610.1243.4865036.0264492631.02674.23934.4013938.4443037.5133719.3281251725.275154.257813107.976588.04155.5280.902283.257349.147348.425323.615113.60073.175113.72471.1015.6055.35652.53314.13674885.40.524737796.8161294.530.350347791.4601287.03778.3051314.000.7321301.0319420.4410129.68321379.91166.909.51269.50.3005623.7890.1753.28825304.1919011.45821.12920.1194.550129.44455.81237830.4314283.29140.14658.73815.9719.88303.243300.1780.07610.3376415.528992005.42019.33144.3133151.2282339.6195702.7968751251.7108606.502604145.569468.83021.3217.312218.034272.733273.039420.04788.54557.95088.55555.9804.8076.79962.44617.19495206.21.154653423.155355.241.512713373.885539.483344.145524.502.736212.405611.7635510.22263141.93119.204.66948.50.7094725.0094.1194.86620846.59023.871188.2189.096299.95507.29597026.0381251.74457.37755.32434.6641.54633.217626.9190.11110.5292023.871261903.91991.33155.3913159.8522350.2130250.4492191812.971609.065104115.832649.34196.5293.551294.381365.547367.942398.404116.63076.149116.57374.2874.9135.67329.0068.22673517.30.398005458.201753.3560.352180459.576746.464462.855744.3550.6686410.5514670.4743113.28160480.79256.287.62176.30.5256920.19284.1871.2297870.046444.24638.12582.7469.94283.401157.9332710.19951599.91326.835134.74014.8018.12551.827544.5840.08945.6774279.648781934.32006.32891.8462892.7192279.8147211.2395831505.180350.648438126.185567.43942.7270.067269.659327.093322.924335.411103.37270.985103.25168.3494.4840.58631.5048.9761754854.013922802.874495.904.233823198.074449.573117.454793.755.987126.924717.328628.70022172.1695.143.49039.81.71710111.14OpenBenchmarking.org

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure HC - Xeon 8168Azure H - Xeon E5-2667 v360120180240300SE +/- 0.23, N = 3SE +/- 0.20, N = 3SE +/- 0.39, N = 3SE +/- 0.02, N = 3SE +/- 3.12, N = 337.8039.2090.1894.12284.191. (CXX) g++ options: -O2 -lOpenCL

GROMACS

Water Benchmark

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v33691215SE +/- 0.009, N = 3SE +/- 0.009, N = 3SE +/- 0.003, N = 3SE +/- 0.025, N = 15SE +/- 0.002, N = 39.0418.4584.8663.2881.2291. (CXX) g++ options: -O2 -pthread -lrt -lpthread -lm

NAS Parallel Benchmarks

Test / Class: LU.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v312K24K36K48K60KSE +/- 428.57, N = 14SE +/- 21.24, N = 3SE +/- 114.01, N = 3SE +/- 33.92, N = 356682.8253829.3425304.197870.041. (F9X) gfortran options: -O3 -march=native -fexceptions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

LULESH

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v39K18K27K36K45KSE +/- 476.10, N = 3SE +/- 44.77, N = 3SE +/- 25.50, N = 3SE +/- 81.47, N = 3SE +/- 28.01, N = 341636.6034803.5420846.5919011.466444.251. (CXX) g++ options: -O3 -fopenmp -lm -fexceptions -pthread -lmpi_cxx -lmpi

SVT-HEVC

Tuning: 1 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v31020304050SE +/- 0.60, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 345.3936.1223.8721.128.121. (CC) gcc options: -fPIE -fPIC -O2 -O3 -pie -rdynamic -lpthread -lrt

Zstd Compression

Compression Level: 8 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Compression SpeedAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v37001400210028003500SE +/- 45.10, N = 3SE +/- 33.03, N = 15SE +/- 6.93, N = 3SE +/- 6.06, N = 3SE +/- 1.83, N = 33184.22653.11188.2920.1582.71. (CC) gcc options: -O3 -pthread -lz -llzma

Timed Node.js Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 15.11Time To CompileAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3100200300400500SE +/- 1.09, N = 3SE +/- 0.83, N = 3SE +/- 0.40, N = 3SE +/- 0.16, N = 3SE +/- 0.95, N = 3111.33118.97189.10194.55469.94

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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v370140210280350SE +/- 4.36, N = 3SE +/- 4.18, N = 3SE +/- 0.20, N = 3SE +/- 1.10, N = 8SE +/- 1.28, N = 15337.71299.95140.59129.4483.401. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -O2 -pie -rdynamic -lpthread -lrt -lm

Xcompact3d Incompact3d

Input: X3D-benchmarking input.i3d

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: X3D-benchmarking input.i3dAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure HC - Xeon 8168Azure H - Xeon E5-2667 v32004006008001000SE +/- 0.17, N = 3SE +/- 0.07, N = 3SE +/- 2.96, N = 3SE +/- 7.11, N = 3SE +/- 12.66, N = 3287.60318.91455.81507.301157.931. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -fexceptions -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

High Performance Conjugate Gradient

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure HC - Xeon 8168Azure H - Xeon E5-2667 v3918273645SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.17, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 839.0637.2730.4326.0410.201. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -fexceptions -pthread -lmpi_cxx -lmpi

Timed LLVM Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3130260390520650SE +/- 1.89, N = 3SE +/- 1.99, N = 3SE +/- 2.31, N = 3SE +/- 3.41, N = 4SE +/- 2.26, N = 3163.76174.74251.74283.29599.91

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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v320406080100SE +/- 0.34, N = 3SE +/- 0.65, N = 3SE +/- 0.24, N = 3SE +/- 0.54, N = 12SE +/- 0.35, N = 394.0378.1057.3840.1526.841. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

Timed Linux Kernel Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3306090120150SE +/- 0.58, N = 15SE +/- 0.58, N = 13SE +/- 0.60, N = 15SE +/- 0.45, N = 13SE +/- 1.07, N = 1242.0144.9255.3258.74134.74

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPUAzure HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3816243240SE +/- 0.19, N = 3SE +/- 0.42, N = 4SE +/- 0.31, N = 15SE +/- 0.21, N = 12SE +/- 0.18, N = 434.6634.5323.1215.9714.80

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUAzure HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3918273645SE +/- 0.22, N = 3SE +/- 0.39, N = 3SE +/- 0.21, N = 3SE +/- 0.21, N = 15SE +/- 0.03, N = 341.5438.3925.5719.8818.12

Botan

Test: ChaCha20Poly1305

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305Azure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 7551140280420560700SE +/- 0.27, N = 3SE +/- 0.30, N = 3SE +/- 0.90, N = 3SE +/- 2.26, N = 3SE +/- 1.39, N = 3667.77633.22615.80551.83303.241. (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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 7551140280420560700SE +/- 0.06, N = 3SE +/- 0.55, N = 3SE +/- 0.78, N = 3SE +/- 1.38, N = 3SE +/- 1.37, N = 3658.37626.92610.86544.58300.181. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 75510.03620.07240.10860.14480.181SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1610.1240.1110.0890.0761. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

Pennant

Test: leblancbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure HC - Xeon 8168Azure H - Xeon E5-2667 v31020304050SE +/- 0.017738, N = 3SE +/- 0.009939, N = 3SE +/- 0.027196, N = 3SE +/- 0.037598, N = 3SE +/- 0.671685, N = 123.3372013.48650310.33764010.52920045.6774201. (CXX) g++ options: -fopenmp -fexceptions -pthread -lmpi_cxx -lmpi

Pennant

Test: sedovbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure HC - Xeon 8168Azure H - Xeon E5-2667 v320406080100SE +/- 0.003905, N = 3SE +/- 0.012548, N = 3SE +/- 0.085157, N = 3SE +/- 0.048781, N = 3SE +/- 0.744226, N = 155.8334196.02644915.52899023.87126079.6487801. (CXX) g++ options: -fopenmp -fexceptions -pthread -lmpi_cxx -lmpi

Zstd Compression

Compression Level: 19 - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression SpeedAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3Azure HC - Xeon 81688001600240032004000SE +/- 6.14, N = 15SE +/- 1.17, N = 15SE +/- 0.66, N = 15SE +/- 2.70, N = 3SE +/- 2.22, N = 153717.52631.02005.41934.31903.91. (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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3Azure HC - Xeon 81688001600240032004000SE +/- 12.70, N = 15SE +/- 2.70, N = 15SE +/- 0.32, N = 12SE +/- 2.28, N = 5SE +/- 2.46, N = 153727.52674.22019.32006.31991.31. (CC) gcc options: -O3 -pthread -lz -llzma

Botan

Test: AES-256

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v312002400360048006000SE +/- 3.73, N = 3SE +/- 2.68, N = 3SE +/- 0.18, N = 3SE +/- 16.79, N = 3SE +/- 4.69, N = 35412.133934.403155.393144.312891.851. (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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v312002400360048006000SE +/- 7.85, N = 3SE +/- 3.26, N = 3SE +/- 0.03, N = 3SE +/- 1.16, N = 3SE +/- 5.16, N = 35407.483938.443159.853151.232892.721. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v39001800270036004500SE +/- 4.89, N = 7SE +/- 4.57, N = 3SE +/- 2.50, N = 3SE +/- 1.19, N = 15SE +/- 13.10, N = 34256.83037.52350.22339.62279.81. (CC) gcc options: -O3 -pthread -lz -llzma

FinanceBench

Benchmark: Bonds OpenMP

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Bonds OpenMPAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755140K80K120K160K200KSE +/- 454.05, N = 3SE +/- 1475.96, N = 4SE +/- 476.22, N = 3SE +/- 150.66, N = 3SE +/- 1191.05, N = 3104884.29130250.45133719.33147211.24195702.801. (CXX) g++ options: -O3 -march=native -fopenmp

QuantLib

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21Azure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 75515001000150020002500SE +/- 4.80, N = 3SE +/- 7.80, N = 3SE +/- 0.77, N = 3SE +/- 4.07, N = 3SE +/- 1.79, N = 32299.31812.91725.21505.11251.71. (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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755120K40K60K80K100KSE +/- 228.72, N = 3SE +/- 15.23, N = 3SE +/- 166.77, N = 3SE +/- 85.66, N = 3SE +/- 324.87, N = 359196.5571609.0775154.2680350.65108606.501. (CXX) g++ options: -O3 -march=native -fopenmp

Rodinia

Test: OpenMP HotSpot3D

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP HotSpot3DAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 7551306090120150SE +/- 1.18, N = 15SE +/- 0.89, N = 3SE +/- 0.21, N = 3SE +/- 0.43, N = 3SE +/- 1.35, N = 382.49107.98115.83126.19145.571. (CXX) g++ options: -O2 -lOpenCL

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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 7551170340510680850SE +/- 6.99, N = 7SE +/- 6.29, N = 3SE +/- 0.81, N = 3SE +/- 1.99, N = 3SE +/- 3.02, N = 15771.9649.3588.0567.4468.81. (CC) gcc options: -O3 -pthread -lz -llzma

GNU GMP GMPbench

Total Time

OpenBenchmarking.orgGMPbench Score, More Is BetterGNU GMP GMPbench 6.2.1Total TimeAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 7551100020003000400050004893.44196.54155.53942.73021.31. (CC) gcc options: -O3 -fomit-frame-pointer -lm

Botan

Test: Twofish

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: TwofishAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755180160240320400SE +/- 1.10, N = 3SE +/- 0.20, N = 3SE +/- 0.15, N = 3SE +/- 0.46, N = 3SE +/- 2.54, N = 4346.32293.55280.90270.07217.311. (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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755180160240320400SE +/- 0.77, N = 3SE +/- 0.13, N = 3SE +/- 0.22, N = 3SE +/- 0.99, N = 3SE +/- 2.43, N = 4345.78294.38283.26269.66218.031. (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 - DecryptAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755190180270360450SE +/- 0.42, N = 3SE +/- 0.12, N = 3SE +/- 0.56, N = 3SE +/- 0.55, N = 3SE +/- 0.08, N = 3425.16365.55349.15327.09272.731. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: Blowfish

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: BlowfishAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755190180270360450SE +/- 0.47, N = 3SE +/- 0.10, N = 3SE +/- 0.42, N = 3SE +/- 0.68, N = 3SE +/- 0.10, N = 3424.74367.94348.43322.92273.041. (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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HC - Xeon 8168Azure HBv1 - EPYC 755190180270360450SE +/- 0.15, N = 3SE +/- 0.19, N = 3SE +/- 0.39, N = 3SE +/- 0.06, N = 3SE +/- 0.16, N = 3272.61323.62335.41398.40420.05MIN: 272.1 / MAX: 273.5MIN: 322.04 / MAX: 326.05MIN: 329.82 / MAX: 368.39MIN: 398 / MAX: 399.03MIN: 418.9 / MAX: 430.091. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O2 -rdynamic -ldl

Botan

Test: CAST-256

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256Azure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 7551306090120150SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 3133.69116.63113.60103.3788.551. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

Botan

Test: KASUMI

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMIAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755120406080100SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.25, N = 3SE +/- 0.00, N = 387.4876.1573.1870.9957.951. (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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 7551306090120150SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.23, N = 3SE +/- 0.03, N = 2133.67116.57113.72103.2588.561. (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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755120406080100SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.16, N = 3SE +/- 0.00, N = 384.0674.2971.1068.3555.981. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3246810SE +/- 0.16, N = 9SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 36.275.604.914.804.48

Mobile Neural Network

Model: inception-v3

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: inception-v3Azure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure H - Xeon E5-2667 v3Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 755120406080100SE +/- 0.24, N = 15SE +/- 0.32, N = 3SE +/- 1.56, N = 12SE +/- 0.95, N = 12SE +/- 0.96, N = 1234.7135.6740.5955.3676.80MIN: 31.09 / MAX: 427.77MIN: 34.52 / MAX: 128.55MIN: 29.94 / MAX: 479.23MIN: 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

Mobile Neural Network

Model: resnet-v2-50

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.3Model: resnet-v2-50Azure HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure H - Xeon E5-2667 v3Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 75511428425670SE +/- 0.12, N = 3SE +/- 0.25, N = 15SE +/- 0.34, N = 12SE +/- 1.18, N = 12SE +/- 1.07, N = 1229.0129.0931.5052.5362.45MIN: 27.94 / MAX: 76.84MIN: 25.79 / MAX: 252.29MIN: 25.12 / MAX: 279.16MIN: 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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure H - Xeon E5-2667 v3Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 755148121620SE +/- 0.113, N = 15SE +/- 0.058, N = 3SE +/- 0.246, N = 12SE +/- 0.657, N = 12SE +/- 0.162, N = 128.2248.2268.97614.13617.194MIN: 5.66 / MAX: 54.17MIN: 7.29 / MAX: 23.89MIN: 6.66 / MAX: 318.77MIN: 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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v340K80K120K160K200KSE +/- 2854.14, N = 15SE +/- 698.77, N = 3SE +/- 1151.94, N = 15SE +/- 1151.23, N = 15SE +/- 987.91, N = 366320.473517.374885.495206.2175485.0

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 HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v30.90311.80622.70933.61244.5155SE +/- 0.004555, N = 15SE +/- 0.004912, N = 4SE +/- 0.004967, N = 3SE +/- 0.003421, N = 3SE +/- 0.564166, N = 120.3980050.4452230.5247371.1546504.013920MIN: 1.1MIN: 2.031. (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: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUAzure HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 75517001400210028003500SE +/- 2.64, N = 3SE +/- 8.06, N = 15SE +/- 6.35, N = 15SE +/- 133.56, N = 15SE +/- 87.86, N = 15458.20540.19796.822802.873423.15MIN: 462.7MIN: 726.18MIN: 1783.32MIN: 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 HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755111002200330044005500SE +/- 4.09, N = 3SE +/- 14.29, N = 15SE +/- 11.10, N = 15SE +/- 240.85, N = 12SE +/- 122.64, N = 15753.36843.101294.534495.905355.24MIN: 736.05MIN: 711.66MIN: 1179.19MIN: 2965.45MIN: 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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v30.95261.90522.85783.81044.763SE +/- 0.000278, N = 3SE +/- 0.006544, N = 12SE +/- 0.011574, N = 15SE +/- 0.088179, N = 15SE +/- 0.797911, N = 150.2555010.3503470.3521801.5127104.233820MIN: 0.57MIN: 1.341. (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 HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 75517001400210028003500SE +/- 1.52, N = 3SE +/- 8.80, N = 15SE +/- 5.73, N = 15SE +/- 215.78, N = 15SE +/- 110.58, N = 12459.58561.77791.463198.073373.88MIN: 471.83MIN: 718.74MIN: 1805.4MIN: 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 HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755112002400360048006000SE +/- 3.40, N = 3SE +/- 102.39, N = 14SE +/- 13.52, N = 15SE +/- 213.65, N = 12SE +/- 103.35, N = 12746.46962.341287.034449.575539.48MIN: 722.13MIN: 1149.25MIN: 2983.77MIN: 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 HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 75517001400210028003500SE +/- 2.05, N = 3SE +/- 8.62, N = 15SE +/- 8.13, N = 3SE +/- 171.94, N = 12SE +/- 113.65, N = 12462.86530.90778.313117.453344.14MIN: 458.96MIN: 738.78MIN: 1801.23MIN: 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 HC - Xeon 8168Azure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 755112002400360048006000SE +/- 4.60, N = 3SE +/- 25.25, N = 12SE +/- 16.54, N = 15SE +/- 239.22, N = 15SE +/- 83.42, N = 13744.36875.171314.004793.755524.50MIN: 727.38MIN: 1140.59MIN: 2947.39MIN: 4658.381. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v31.34712.69424.04135.38846.7355SE +/- 0.002442, N = 3SE +/- 0.006150, N = 15SE +/- 0.002363, N = 3SE +/- 0.005596, N = 3SE +/- 1.017585, N = 150.4556820.6686410.7321302.7362105.987120MIN: 2.7MIN: 3.171. (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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3246810SE +/- 0.004961, N = 3SE +/- 0.003751, N = 3SE +/- 0.030825, N = 15SE +/- 0.045752, N = 12SE +/- 1.264069, N = 150.4066860.5514671.0319422.4056106.924710MIN: 1.96MIN: 2.21. (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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v3246810SE +/- 0.009558, N = 12SE +/- 0.003995, N = 7SE +/- 0.017521, N = 12SE +/- 0.040927, N = 15SE +/- 2.687031, N = 120.3838550.4410120.4743111.7635507.328620MIN: 0.89MIN: 1.371. (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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure H - Xeon E5-2667 v3Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 75513691215SE +/- 0.02162, N = 3SE +/- 0.00454, N = 3SE +/- 1.08345, N = 15SE +/- 0.28898, N = 15SE +/- 0.43651, N = 153.212043.281608.700229.6832110.22263MIN: 2.58MIN: 3.2MIN: 5.33MIN: 5.56MIN: 4.11. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -O2 -pie -lpthread -ldl

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure H - Xeon E5-2667 v3Azure HBv1 - EPYC 7551120240360480600SE +/- 7.76, N = 3SE +/- 2.25, N = 3SE +/- 21.86, N = 15SE +/- 1.74, N = 6SE +/- 2.68, N = 15548.33480.79379.91172.16141.931. (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 HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v380160240320400SE +/- 4.22, N = 3SE +/- 1.61, N = 3SE +/- 6.53, N = 12SE +/- 1.15, N = 3SE +/- 1.35, N = 3378.01256.28166.90119.2095.141. (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 HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v33691215SE +/- 0.031, N = 3SE +/- 0.218, N = 15SE +/- 0.023, N = 3SE +/- 0.034, N = 11SE +/- 0.022, N = 312.2759.5127.6214.6693.4901. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

Zstd Compression

Compression Level: 19 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression SpeedAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v320406080100SE +/- 0.71, N = 15SE +/- 1.01, N = 15SE +/- 1.18, N = 15SE +/- 0.83, N = 15SE +/- 0.57, N = 378.276.369.548.539.81. (CC) gcc options: -O3 -pthread -lz -llzma

NAMD

ATPase Simulation - 327,506 Atoms

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsAzure HBv3 - EPYC 7V13Azure HBv2 - EPYC 7V12Azure HC - Xeon 8168Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v30.38630.77261.15891.54521.9315SE +/- 0.00027, N = 3SE +/- 0.00059, N = 3SE +/- 0.00126, N = 3SE +/- 0.00507, N = 3SE +/- 0.05413, N = 150.275660.300560.525690.709471.71710

CloverLeaf

Lagrangian-Eulerian Hydrodynamics

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsAzure HBv3 - EPYC 7V13Azure HC - Xeon 8168Azure HBv2 - EPYC 7V12Azure HBv1 - EPYC 7551Azure H - Xeon E5-2667 v320406080100SE +/- 0.83, N = 15SE +/- 0.26, N = 3SE +/- 0.45, N = 12SE +/- 0.11, N = 3SE +/- 1.93, N = 1416.6620.1923.7825.00111.141. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

Phoronix Test Suite v10.8.4