PlaidML Benchmark First Run TR 2990WX + RX Vega AMD Ryzen Threadripper 2990WX 32-Core testing with a ASUS ROG ZENITH EXTREME (1601 BIOS) and AMD Radeon RX 64 8GB on Ubuntu 18.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/1901106-PTS-PLAIDMLB64 .
PlaidML Benchmark First Run TR 2990WX + RX Vega Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution TR 2990WX + RX Vega AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads) ASUS ROG ZENITH EXTREME (1601 BIOS) AMD Family 17h 32768MB Samsung SSD 970 EVO 500GB AMD Radeon RX 64 8GB (1590/800MHz) Realtek ALC1220 ASUS VP28U Intel I211 Gigabit + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad Ubuntu 18.10 4.19.0-041900-generic (x86_64) GNOME Shell 3.30.1 X Server 1.20.1 amdgpu 18.1.0 4.5 Mesa 18.2.2 (LLVM 7.0.0) GCC 8.2.0 ext4 3840x2160 OpenBenchmarking.org - Scaling Governor: acpi-cpufreq ondemand - GLAMOR - Python 2.7.15+ + Python 3.6.7 - __user pointer sanitization + Full AMD retpoline IBPB + SSB disabled via prctl and seccomp
PlaidML Benchmark First Run TR 2990WX + RX Vega plaidml: No - VGG16 - CPU plaidml: No - VGG19 - CPU plaidml: Yes - VGG16 - CPU plaidml: Yes - VGG19 - CPU plaidml: No - VGG16 - OpenCL plaidml: No - VGG19 - OpenCL plaidml: No - IMDB LSTM - CPU plaidml: No - Mobilenet - CPU plaidml: No - ResNet 50 - CPU plaidml: Yes - VGG16 - OpenCL plaidml: Yes - VGG19 - OpenCL plaidml: No - DenseNet 201 - CPU plaidml: No - IMDB LSTM - OpenCL plaidml: No - Mobilenet - OpenCL plaidml: No - NASNer Large - CPU plaidml: No - ResNet 50 - OpenCL plaidml: Yes - Mobilenet - OpenCL plaidml: Yes - ResNet 50 - OpenCL plaidml: No - DenseNet 201 - OpenCL plaidml: No - Inception V3 - OpenCL plaidml: No - NASNer Large - OpenCL plaidml: Yes - DenseNet 201 - OpenCL plaidml: Yes - Inception V3 - OpenCL plaidml: Yes - NASNer Large - OpenCL TR 2990WX + RX Vega 3.24 2.90 3.58 3.09 107.44 93.51 2.38 31.50 8.70 106.19 90.35 2.38 239.92 474.23 0.83 202.24 480.12 191.95 95.76 101.99 31.18 95.62 105.76 29.56 OpenBenchmarking.org
PlaidML FP16: No - Network: VGG16 - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: VGG16 - Device: CPU TR 2990WX + RX Vega 0.729 1.458 2.187 2.916 3.645 3.24
PlaidML FP16: No - Network: VGG19 - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: VGG19 - Device: CPU TR 2990WX + RX Vega 0.6525 1.305 1.9575 2.61 3.2625 2.90
PlaidML FP16: Yes - Network: VGG16 - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: VGG16 - Device: CPU TR 2990WX + RX Vega 0.8055 1.611 2.4165 3.222 4.0275 3.58
PlaidML FP16: Yes - Network: VGG19 - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: VGG19 - Device: CPU TR 2990WX + RX Vega 0.6953 1.3906 2.0859 2.7812 3.4765 3.09
PlaidML FP16: No - Network: VGG16 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: VGG16 - Device: OpenCL TR 2990WX + RX Vega 20 40 60 80 100 107.44
PlaidML FP16: No - Network: VGG19 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: VGG19 - Device: OpenCL TR 2990WX + RX Vega 20 40 60 80 100 93.51
PlaidML FP16: No - Network: IMDB LSTM - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: IMDB LSTM - Device: CPU TR 2990WX + RX Vega 0.5355 1.071 1.6065 2.142 2.6775 2.38
PlaidML FP16: No - Network: Mobilenet - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: Mobilenet - Device: CPU TR 2990WX + RX Vega 7 14 21 28 35 31.50
PlaidML FP16: No - Network: ResNet 50 - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: ResNet 50 - Device: CPU TR 2990WX + RX Vega 2 4 6 8 10 8.70
PlaidML FP16: Yes - Network: VGG16 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: VGG16 - Device: OpenCL TR 2990WX + RX Vega 20 40 60 80 100 106.19
PlaidML FP16: Yes - Network: VGG19 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: VGG19 - Device: OpenCL TR 2990WX + RX Vega 20 40 60 80 100 90.35
PlaidML FP16: No - Network: DenseNet 201 - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: DenseNet 201 - Device: CPU TR 2990WX + RX Vega 0.5355 1.071 1.6065 2.142 2.6775 2.38
PlaidML FP16: No - Network: IMDB LSTM - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: IMDB LSTM - Device: OpenCL TR 2990WX + RX Vega 50 100 150 200 250 239.92
PlaidML FP16: No - Network: Mobilenet - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: Mobilenet - Device: OpenCL TR 2990WX + RX Vega 100 200 300 400 500 474.23
PlaidML FP16: No - Network: NASNer Large - Device: CPU OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: NASNer Large - Device: CPU TR 2990WX + RX Vega 0.1868 0.3736 0.5604 0.7472 0.934 0.83
PlaidML FP16: No - Network: ResNet 50 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: ResNet 50 - Device: OpenCL TR 2990WX + RX Vega 40 80 120 160 200 202.24
PlaidML FP16: Yes - Network: Mobilenet - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: Mobilenet - Device: OpenCL TR 2990WX + RX Vega 100 200 300 400 500 480.12
PlaidML FP16: Yes - Network: ResNet 50 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: ResNet 50 - Device: OpenCL TR 2990WX + RX Vega 40 80 120 160 200 191.95
PlaidML FP16: No - Network: DenseNet 201 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: DenseNet 201 - Device: OpenCL TR 2990WX + RX Vega 20 40 60 80 100 95.76
PlaidML FP16: No - Network: Inception V3 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: Inception V3 - Device: OpenCL TR 2990WX + RX Vega 20 40 60 80 100 101.99
PlaidML FP16: No - Network: NASNer Large - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: No - Network: NASNer Large - Device: OpenCL TR 2990WX + RX Vega 7 14 21 28 35 31.18
PlaidML FP16: Yes - Network: DenseNet 201 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: DenseNet 201 - Device: OpenCL TR 2990WX + RX Vega 20 40 60 80 100 95.62
PlaidML FP16: Yes - Network: Inception V3 - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: Inception V3 - Device: OpenCL TR 2990WX + RX Vega 20 40 60 80 100 105.76
PlaidML FP16: Yes - Network: NASNer Large - Device: OpenCL OpenBenchmarking.org Examples Per Second, More Is Better PlaidML FP16: Yes - Network: NASNer Large - Device: OpenCL TR 2990WX + RX Vega 7 14 21 28 35 29.56
Phoronix Test Suite v10.8.4