ripper-pytorch AMD Ryzen Threadripper 7960X 24-Cores testing with a ASRock TRX50 WS (7.09 BIOS) and NVIDIA GeForce RTX 4060 Ti 16GB on Ubuntu 22.04 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2403309-NE-RIPPERPYT22 ripper-pytorch Processor: AMD Ryzen Threadripper 7960X 24-Cores @ 8.23GHz (24 Cores / 48 Threads), Motherboard: ASRock TRX50 WS (7.09 BIOS), Chipset: AMD Device 14a4, Memory: 128GB, Disk: 2000GB Samsung SSD 980 PRO with Heatsink 2TB, Graphics: NVIDIA GeForce RTX 4060 Ti 16GB, Audio: AMD Device 14cc, Monitor: SyncMaster, Network: Aquantia Device 04c0 + Realtek RTL8125 2.5GbE + MEDIATEK Device 0616
OS: Ubuntu 22.04, Kernel: 6.5.0-26-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server 1.21.1.4, Display Driver: NVIDIA 545.29.06, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.3.99, Vulkan: 1.3.260, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 1680x1050
Kernel Notes: Transparent Huge Pages: madviseProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: performance) - CPU Microcode: 0xa108105Python Notes: Python 3.10.12Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 1 - Model: ResNet-50 ripper-pytorch 14 28 42 56 70 SE +/- 0.27, N = 3 64.47 MIN: 49.55 / MAX: 66.67
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 1 - Model: ResNet-152 ripper-pytorch 6 12 18 24 30 SE +/- 0.15, N = 3 25.21 MIN: 20.13 / MAX: 25.88
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 16 - Model: ResNet-50 ripper-pytorch 11 22 33 44 55 SE +/- 0.11, N = 3 49.48 MIN: 46.91 / MAX: 50.12
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 32 - Model: ResNet-50 ripper-pytorch 11 22 33 44 55 SE +/- 0.31, N = 3 49.00 MIN: 45.84 / MAX: 50.07
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 64 - Model: ResNet-50 ripper-pytorch 11 22 33 44 55 SE +/- 0.26, N = 3 49.02 MIN: 46.16 / MAX: 49.85
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 16 - Model: ResNet-152 ripper-pytorch 5 10 15 20 25 SE +/- 0.12, N = 3 19.16 MIN: 17.87 / MAX: 19.55
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 ripper-pytorch 11 22 33 44 55 SE +/- 0.22, N = 3 49.30 MIN: 46.71 / MAX: 50.14
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 32 - Model: ResNet-152 ripper-pytorch 5 10 15 20 25 SE +/- 0.02, N = 3 19.30 MIN: 18.88 / MAX: 19.46
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 ripper-pytorch 11 22 33 44 55 SE +/- 0.17, N = 3 48.95 MIN: 45.69 / MAX: 49.93
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 64 - Model: ResNet-152 ripper-pytorch 5 10 15 20 25 SE +/- 0.02, N = 3 19.27 MIN: 18.8 / MAX: 19.46
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-152 ripper-pytorch 5 10 15 20 25 SE +/- 0.02, N = 3 19.26 MIN: 18.76 / MAX: 19.4
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: ResNet-152 ripper-pytorch 5 10 15 20 25 SE +/- 0.07, N = 3 19.20 MIN: 18.72 / MAX: 19.42
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l ripper-pytorch 4 8 12 16 20 SE +/- 0.07, N = 3 15.20 MIN: 12.84 / MAX: 15.51
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l ripper-pytorch 3 6 9 12 15 SE +/- 0.01, N = 3 10.82 MIN: 9.7 / MAX: 11.03
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l ripper-pytorch 3 6 9 12 15 SE +/- 0.01, N = 3 10.82 MIN: 9.47 / MAX: 11.07
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_l ripper-pytorch 3 6 9 12 15 SE +/- 0.07, N = 3 10.77 MIN: 9.05 / MAX: 11.14
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l ripper-pytorch 3 6 9 12 15 SE +/- 0.04, N = 3 10.92 MIN: 9.71 / MAX: 11.17
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l ripper-pytorch 3 6 9 12 15 SE +/- 0.02, N = 3 10.82 MIN: 9.59 / MAX: 11.05
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50 ripper-pytorch 70 140 210 280 350 SE +/- 2.80, N = 3 334.53 MIN: 278.54 / MAX: 343.94
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152 ripper-pytorch 30 60 90 120 150 SE +/- 0.39, N = 3 118.70 MIN: 104.59 / MAX: 121.02
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50 ripper-pytorch 70 140 210 280 350 SE +/- 1.97, N = 3 308.84 MIN: 286.79 / MAX: 316.03
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50 ripper-pytorch 70 140 210 280 350 SE +/- 2.54, N = 3 313.80 MIN: 289.68 / MAX: 320.89
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50 ripper-pytorch 70 140 210 280 350 SE +/- 3.16, N = 3 310.95 MIN: 284.77 / MAX: 319.52
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152 ripper-pytorch 30 60 90 120 150 SE +/- 1.27, N = 3 118.70 MIN: 105.17 / MAX: 122.87
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50 ripper-pytorch 70 140 210 280 350 SE +/- 1.84, N = 3 316.14 MIN: 296.19 / MAX: 323.31
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152 ripper-pytorch 30 60 90 120 150 SE +/- 0.31, N = 3 119.37 MIN: 106.4 / MAX: 120.94
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50 ripper-pytorch 70 140 210 280 350 SE +/- 1.73, N = 3 316.76 MIN: 288.88 / MAX: 324.66
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152 ripper-pytorch 30 60 90 120 150 SE +/- 0.89, N = 3 118.01 MIN: 105.65 / MAX: 120.3
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152 ripper-pytorch 30 60 90 120 150 SE +/- 1.01, N = 3 120.96 MIN: 107.66 / MAX: 123.82
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152 ripper-pytorch 30 60 90 120 150 SE +/- 0.12, N = 3 120.45 MIN: 107.48 / MAX: 122.04
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l ripper-pytorch 14 28 42 56 70 SE +/- 0.30, N = 3 62.79 MIN: 54.78 / MAX: 63.93
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l ripper-pytorch 14 28 42 56 70 SE +/- 0.18, N = 3 61.22 MIN: 53.84 / MAX: 62.11
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l ripper-pytorch 14 28 42 56 70 SE +/- 0.60, N = 3 61.10 MIN: 54.19 / MAX: 62.82
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l ripper-pytorch 14 28 42 56 70 SE +/- 0.60, N = 5 61.07 MIN: 52.92 / MAX: 62.55
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l ripper-pytorch 13 26 39 52 65 SE +/- 0.52, N = 3 60.00 MIN: 52.58 / MAX: 61.37
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l ripper-pytorch 13 26 39 52 65 SE +/- 0.62, N = 3 60.24 MIN: 53.1 / MAX: 61.93
ripper-pytorch Processor: AMD Ryzen Threadripper 7960X 24-Cores @ 8.23GHz (24 Cores / 48 Threads), Motherboard: ASRock TRX50 WS (7.09 BIOS), Chipset: AMD Device 14a4, Memory: 128GB, Disk: 2000GB Samsung SSD 980 PRO with Heatsink 2TB, Graphics: NVIDIA GeForce RTX 4060 Ti 16GB, Audio: AMD Device 14cc, Monitor: SyncMaster, Network: Aquantia Device 04c0 + Realtek RTL8125 2.5GbE + MEDIATEK Device 0616
OS: Ubuntu 22.04, Kernel: 6.5.0-26-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server 1.21.1.4, Display Driver: NVIDIA 545.29.06, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.3.99, Vulkan: 1.3.260, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 1680x1050
Kernel Notes: Transparent Huge Pages: madviseProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: performance) - CPU Microcode: 0xa108105Python Notes: Python 3.10.12Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: 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
Testing initiated at 30 March 2024 12:24 by user andrew.