mkl haswell Intel Xeon E5-1680 v3 testing with a ASUS X99-A (3902 BIOS) and eVGA NVIDIA NVE7 1GB on Ubuntu 18.04 via the Phoronix Test Suite. Intel Xeon E5-1680 v3: Processor: Intel Xeon E5-1680 v3 @ 3.80GHz (8 Cores / 16 Threads), Motherboard: ASUS X99-A (3902 BIOS), Chipset: Intel Xeon E7 v3/Xeon, Memory: 16384MB, Disk: 60GB Patriot Torch, Graphics: eVGA NVIDIA NVE7 1GB, Audio: Realtek ALC1150, Monitor: VE228, Network: Intel I218-V OS: Ubuntu 18.04, Kernel: 4.18.0-20-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: modesetting 1.20.1, OpenGL: 4.3 Mesa 18.2.8, Compiler: GCC 7.4.0, File-System: ext4, Screen Resolution: 1920x1080 MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: f32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 11.89 |================================================ MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: f32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 156.85 |=============================================== MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: u8s8u8s32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 8.32 |================================================= MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: u8s8f32s32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 7.34 |================================================= MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: u8s8u8s32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 106.63 |=============================================== MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: u8s8f32s32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 103.53 |=============================================== MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: f32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 23.36 |================================================ MKL-DNN 2019-04-16 Harness: Convolution Batch conv_all - Data Type: f32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 3388.48 |============================================== MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_1d - Data Type: f32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 7.15 |================================================= MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_3d - Data Type: f32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 9.35 |================================================= MKL-DNN 2019-04-16 Harness: Convolution Batch conv_alexnet - Data Type: f32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 447.17 |=============================================== MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_all - Data Type: f32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 2865.70 |============================================== MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: u8s8u8s32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 14458.40 |============================================= MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: u8s8f32s32 ms < Lower Is Better Intel Xeon E5-1680 v3 . 14471.60 |=============================================