pytorch 2.2.1 AMD EPYC Siena

AMD EPYC 8534P 64-Core testing with a AMD Cinnabar (RCB1009C BIOS) and ASPEED on Ubuntu 23.10 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2403269-NE-PYTORCH2202&grw&rdt.

pytorch 2.2.1 AMD EPYC SienaProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionabcAMD EPYC 8534P 64-Core @ 2.30GHz (64 Cores / 128 Threads)AMD Cinnabar (RCB1009C BIOS)AMD Device 14a46 x 32GB DRAM-4800MT/s Samsung M321R4GA0BB0-CQKMG3201GB Micron_7450_MTFDKCB3T2TFS + 2000GB Corsair MP700ASPEED2 x Broadcom NetXtreme BCM5720 PCIeUbuntu 23.106.8.1-060801-generic (x86_64)GNOME Shell 45.2X Server 1.21.1.7GCC 13.2.0ext41920x1200OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseProcessor Details- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212 Python Details- Python 3.11.6Security Details- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

pytorch 2.2.1 AMD EPYC Sienapytorch: CPU - 1 - ResNet-50pytorch: CPU - 1 - ResNet-152pytorch: CPU - 16 - ResNet-50pytorch: CPU - 32 - ResNet-50pytorch: CPU - 64 - ResNet-50pytorch: CPU - 16 - ResNet-152pytorch: CPU - 256 - ResNet-50pytorch: CPU - 32 - ResNet-152pytorch: CPU - 512 - ResNet-50pytorch: CPU - 64 - ResNet-152pytorch: CPU - 256 - ResNet-152pytorch: CPU - 512 - ResNet-152pytorch: CPU - 1 - Efficientnet_v2_lpytorch: CPU - 16 - Efficientnet_v2_lpytorch: CPU - 32 - Efficientnet_v2_lpytorch: CPU - 64 - Efficientnet_v2_lpytorch: CPU - 256 - Efficientnet_v2_lpytorch: CPU - 512 - Efficientnet_v2_labc41.1115.1533.3032.9433.8613.1933.6713.0733.4212.9612.9713.248.645.655.645.675.655.6840.6714.8833.3133.6533.5113.0633.3413.0333.5112.9913.1013.098.785.655.625.635.675.6540.7515.0333.5433.3633.6513.1033.5412.9933.6213.2213.3913.108.645.625.695.665.685.65OpenBenchmarking.org

PyTorch

Device: CPU - Batch Size: 1 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-50abc91827364541.1140.6740.75MIN: 39.95 / MAX: 41.96MIN: 39.57 / MAX: 41.58MIN: 39.84 / MAX: 41.56

PyTorch

Device: CPU - Batch Size: 1 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-152abc4812162015.1514.8815.03MIN: 14.65 / MAX: 15.33MIN: 14.71 / MAX: 15MIN: 9.88 / MAX: 15.14

PyTorch

Device: CPU - Batch Size: 16 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-50abc81624324033.3033.3133.54MIN: 16.55 / MAX: 33.78MIN: 32.5 / MAX: 33.78MIN: 18.28 / MAX: 34

PyTorch

Device: CPU - Batch Size: 32 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: ResNet-50abc81624324032.9433.6533.36MIN: 14.77 / MAX: 33.36MIN: 32.2 / MAX: 34.12MIN: 31.61 / MAX: 33.8

PyTorch

Device: CPU - Batch Size: 64 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-50abc81624324033.8633.5133.65MIN: 32.86 / MAX: 34.37MIN: 15.58 / MAX: 33.94MIN: 32.6 / MAX: 34.13

PyTorch

Device: CPU - Batch Size: 16 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-152abc369121513.1913.0613.10MIN: 13.07 / MAX: 13.3MIN: 8.77 / MAX: 13.16MIN: 12.7 / MAX: 13.21

PyTorch

Device: CPU - Batch Size: 256 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-50abc81624324033.6733.3433.54MIN: 32.41 / MAX: 34.14MIN: 32.81 / MAX: 33.77MIN: 16.1 / MAX: 34.1

PyTorch

Device: CPU - Batch Size: 32 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: ResNet-152abc369121513.0713.0312.99MIN: 8.78 / MAX: 13.2MIN: 12.7 / MAX: 13.14MIN: 8.89 / MAX: 13.09

PyTorch

Device: CPU - Batch Size: 512 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-50abc81624324033.4233.5133.62MIN: 32.29 / MAX: 33.81MIN: 32.71 / MAX: 33.92MIN: 32.63 / MAX: 34.14

PyTorch

Device: CPU - Batch Size: 64 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-152abc369121512.9612.9913.22MIN: 12.23 / MAX: 13.06MIN: 12.86 / MAX: 13.09MIN: 13.02 / MAX: 13.32

PyTorch

Device: CPU - Batch Size: 256 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-152abc369121512.9713.1013.39MIN: 8.8 / MAX: 13.09MIN: 13 / MAX: 13.2MIN: 8.95 / MAX: 13.48

PyTorch

Device: CPU - Batch Size: 512 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-152abc369121513.2413.0913.10MIN: 12.34 / MAX: 13.35MIN: 8.81 / MAX: 13.18MIN: 12.96 / MAX: 13.2

PyTorch

Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_labc2468108.648.788.64MIN: 8.55 / MAX: 8.71MIN: 6.74 / MAX: 8.86MIN: 8.5 / MAX: 8.72

PyTorch

Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_labc1.27132.54263.81395.08526.35655.655.655.62MIN: 4.48 / MAX: 5.76MIN: 4.71 / MAX: 5.76MIN: 4.63 / MAX: 5.72

PyTorch

Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_labc1.28032.56063.84095.12126.40155.645.625.69MIN: 4.7 / MAX: 5.75MIN: 5.06 / MAX: 5.71MIN: 4.72 / MAX: 5.78

PyTorch

Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_labc1.27582.55163.82745.10326.3795.675.635.66MIN: 5.32 / MAX: 5.76MIN: 4.88 / MAX: 5.72MIN: 4.52 / MAX: 5.76

PyTorch

Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_labc1.2782.5563.8345.1126.395.655.675.68MIN: 4.62 / MAX: 5.73MIN: 4.55 / MAX: 5.76MIN: 5.4 / MAX: 5.78

PyTorch

Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_labc1.2782.5563.8345.1126.395.685.655.65MIN: 4.6 / MAX: 5.77MIN: 5.38 / MAX: 5.73MIN: 4.41 / MAX: 5.74

Phoronix Test Suite v10.8.5