pytorch-gpu-defaultconfig

VMware testing on Ubuntu 22.04.4 LTS via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2405219-NE-PYTORCHGP66.

pytorch-gpu-defaultconfigProcessorMotherboardMemoryDiskGraphicsOSKernelDisplay DriverCompilerFile-SystemScreen ResolutionSystem Layer12 x Intel Xeon Gold 6444Y - NVIDIA L40S-8C 8GB -12 x Intel Xeon Gold 6444Y (23 Cores)Intel 440BX (VMW71.00V.21100432.B64.2301110304 BIOS)240GB268GB Virtual diskNVIDIA L40S-8C 8GBUbuntu 22.04.4 LTS5.14.0-284.64.1.el9_2.x86_64 (x86_64)NVIDIAGCC 11.4.0nfs1024x768VMwareOpenBenchmarking.org- Transparent Huge Pages: always- CPU Microcode: 0x2b000571- Python 3.10.12- gather_data_sampling: Not affected + itlb_multihit: KVM: Mitigation of VMX unsupported + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Unknown: No mitigations + retbleed: Not affected + spec_rstack_overflow: Not affected + 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 IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

pytorch-gpu-defaultconfigpytorch: NVIDIA CUDA GPU - 1 - ResNet-50pytorch: NVIDIA CUDA GPU - 1 - ResNet-152pytorch: NVIDIA CUDA GPU - 512 - ResNet-50pytorch: NVIDIA CUDA GPU - 512 - ResNet-152pytorch: NVIDIA CUDA GPU - 1 - Efficientnet_v2_lpytorch: NVIDIA CUDA GPU - 512 - Efficientnet_v2_l12 x Intel Xeon Gold 6444Y - NVIDIA L40S-8C 8GB -327.16119.95316.59120.0761.1261.86OpenBenchmarking.org

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-5012 x Intel Xeon Gold 6444Y - NVIDIA L40S-8C 8GB -70140210280350SE +/- 0.56, N = 3327.16MIN: 193.7 / MAX: 336.22

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-15212 x Intel Xeon Gold 6444Y - NVIDIA L40S-8C 8GB -306090120150SE +/- 0.72, N = 3119.95MIN: 89.65 / MAX: 122.3

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-5012 x Intel Xeon Gold 6444Y - NVIDIA L40S-8C 8GB -70140210280350SE +/- 1.05, N = 3316.59MIN: 186.55 / MAX: 323.06

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-15212 x Intel Xeon Gold 6444Y - NVIDIA L40S-8C 8GB -306090120150SE +/- 1.17, N = 3120.07MIN: 104 / MAX: 122.75

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l12 x Intel Xeon Gold 6444Y - NVIDIA L40S-8C 8GB -1428425670SE +/- 0.21, N = 361.12MIN: 53.05 / MAX: 62.95

PyTorch

Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l12 x Intel Xeon Gold 6444Y - NVIDIA L40S-8C 8GB -1428425670SE +/- 0.48, N = 361.86MIN: 53.21 / MAX: 63.33

Phoronix Test Suite v10.8.4