BOOST - BIS

BOOST

HTML result view exported from: https://openbenchmarking.org/result/2406285-NE-BOOSTBIS227&grs.

BOOST - BISProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDisplay DriverVulkanCompilerFile-SystemScreen ResolutionSystem LayerBOOSTIntel Xeon E5-2667 v3 (4 Cores / 8 Threads)Blade Shadow ShadowM v2.0 (1.1.3 BIOS)Intel 82G33/G31/P35/P31 + ICH91 x 12GB RAM-2400MT/s Blade R5NVNF8QVGPTY1-26C215GB QEMU HDDRed Hat QXL paravirtual graphic card 8GBRed Hat Virtio deviceUbuntu 22.045.15.0-113-generic (x86_64)NVIDIA1.3.277GCC 11.4.0ext41280x800KVMOpenBenchmarking.org- Transparent Huge Pages: madvise- CPU Microcode: 0x49- Python 3.10.12- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Mitigation of PTE Inversion + mds: Mitigation of Clear buffers; SMT Host state unknown + meltdown: Mitigation of PTI + mmio_stale_data: Mitigation of Clear buffers; SMT Host state unknown + retbleed: Not affected + spec_rstack_overflow: 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 Retpolines; IBPB: conditional; IBRS_FW; STIBP: conditional; RSB filling; PBRSB-eIBRS: Not affected; BHI: Retpoline + srbds: Not affected + tsx_async_abort: Not affected

BOOST - BISpytorch: CPU - 16 - Efficientnet_v2_lpytorch: CPU - 16 - ResNet-152pytorch: CPU - 16 - ResNet-50BOOST3.074.3010.32OpenBenchmarking.org

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_lBOOST0.69081.38162.07242.76323.454SE +/- 0.02, N = 33.07MIN: 2.88 / MAX: 3.14

PyTorch

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

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-152BOOST0.96751.9352.90253.874.8375SE +/- 0.05, N = 34.30MIN: 3.22 / MAX: 4.52

PyTorch

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

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-50BOOST3691215SE +/- 0.10, N = 610.32MIN: 6.64 / MAX: 10.93

Phoronix Test Suite v10.8.5