CPU PlaidML Linux Benchmarks

Intel Core i9-7980XE testing with a ASUS PRIME X299-A (1602 BIOS) and NVIDIA NV120 12GB on Fedora 29 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/1901129-SP-1901119SP76&sor&grs.

CPU PlaidML Linux BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionUbuntu 18.10openSUSE TumbleweedFedora Workstation 29Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads)ASUS PRIME X299-A (1602 BIOS)Intel Sky Lake-E DMI3 Registers16384MBSamsung SSD 970 EVO 500GBNVIDIA NV120 12GBRealtek ALC1220ASUS PB278Intel I219-VUbuntu 18.104.18.0-13-generic (x86_64)GNOME Shell 3.30.1X Server 1.20.1modesetting 1.20.14.3 Mesa 18.2.2GCC 8.2.0ext42560x1440NVIDIA GeForce GTX TITAN X 12GBopenSUSE Tumbleweed 201901084.19.12-1-default (x86_64)KDE PlasmaX Server 1.20.3modesetting 1.20.34.3 Mesa 18.3.1btrfsNVIDIA NV120 12GBFedora 294.19.13-300.fc29.x86_64 (x86_64)GNOME Shell 3.30.2X Server + Wayland4.3 Mesa 18.2.8GCC 8.2.1 20181215ext4OpenBenchmarking.orgProcessor Details- Scaling Governor: intel_pstate powersavePython Details- Ubuntu 18.10: Python 2.7.15+ + Python 3.6.7- openSUSE Tumbleweed: Python 2.7.15 + Python 3.6.5- Fedora Workstation 29: Python 2.7.15 + Python 3.7.2Security Details- Ubuntu 18.10: KPTI + __user pointer sanitization + Full generic retpoline IBPB IBRS_FW + SSB disabled via prctl and seccomp + PTE Inversion; VMX: conditional cache flushes SMT vulnerable - openSUSE Tumbleweed: KPTI + __user pointer sanitization + Indirect Branch Restricted Speculation IBPB: conditional IBRS_FW STIBP: conditional RSB filling + SSB disabled via prctl and seccomp + PTE Inversion; VMX: conditional cache flushes SMT vulnerable - Fedora Workstation 29: SELinux + KPTI + __user pointer sanitization + Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + SSB disabled via prctl and seccomp + PTE Inversion; VMX: conditional cache flushes SMT vulnerable Graphics Details- openSUSE Tumbleweed: EXA

CPU PlaidML Linux Benchmarksplaidml: No - Inference - ResNet 50 - CPUplaidml: No - Inference - VGG19 - CPUplaidml: No - Inference - VGG16 - CPUplaidml: No - Inference - NASNer Large - CPUplaidml: No - Training - VGG19 - CPUplaidml: No - Training - VGG16 - CPUplaidml: Yes - Inference - VGG19 - CPUplaidml: Yes - Inference - VGG16 - CPUplaidml: No - Inference - DenseNet 201 - CPUplaidml: No - Training - IMDB LSTM - CPUplaidml: No - Inference - IMDB LSTM - CPUplaidml: No - Inference - Mobilenet - CPUUbuntu 18.10openSUSE TumbleweedFedora Workstation 298.244.865.731.161.001.233.493.932.712.172.5226.0412.066.357.451.411.211.484.154.582.872.082.3846.519.595.236.161.241.031.273.754.203.142.282.5930.12OpenBenchmarking.org

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.103691215SE +/- 0.12, N = 2SE +/- 0.02, N = 3SE +/- 0.00, N = 212.069.598.24

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.10246810SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 36.355.234.86

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.10246810SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 37.456.165.73

PlaidML

FP16: No - Mode: Inference - Network: NASNer Large - Device: CPU

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: NASNer Large - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.100.31730.63460.95191.26921.5865SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.411.241.16

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: VGG19 - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.100.27230.54460.81691.08921.3615SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.211.031.00

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: VGG16 - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.100.3330.6660.9991.3321.665SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.481.271.23

PlaidML

FP16: Yes - Mode: Inference - Network: VGG19 - Device: CPU

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: VGG19 - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.100.93381.86762.80143.73524.669SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 34.153.753.49

PlaidML

FP16: Yes - Mode: Inference - Network: VGG16 - Device: CPU

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: VGG16 - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.101.03052.0613.09154.1225.1525SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 34.584.203.93

PlaidML

FP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPU

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPUFedora Workstation 29openSUSE TumbleweedUbuntu 18.100.70651.4132.11952.8263.5325SE +/- 0.00, N = 3SE +/- 0.05, N = 9SE +/- 0.00, N = 33.142.872.71

PlaidML

FP16: No - Mode: Training - Network: IMDB LSTM - Device: CPU

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: IMDB LSTM - Device: CPUFedora Workstation 29Ubuntu 18.10openSUSE Tumbleweed0.5131.0261.5392.0522.565SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 32.282.172.08

PlaidML

FP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPU

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPUFedora Workstation 29Ubuntu 18.10openSUSE Tumbleweed0.58281.16561.74842.33122.914SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 32.592.522.38

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: CPUopenSUSE TumbleweedFedora Workstation 29Ubuntu 18.101122334455SE +/- 1.08, N = 12SE +/- 0.03, N = 3SE +/- 0.10, N = 346.5130.1226.04

Phoronix Test Suite v10.8.5