xeon april silver Intel Xeon Silver 4216 testing with a TYAN S7100AG2NR (V4.02 BIOS) and ASPEED on Debian 11 via the Phoronix Test Suite. A: Processor: Intel Xeon Silver 4216 @ 3.20GHz (16 Cores / 32 Threads), Motherboard: TYAN S7100AG2NR (V4.02 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 46GB, Disk: 240GB Corsair Force MP500, Graphics: ASPEED, Audio: Realtek ALC892, Network: 2 x Intel I350 OS: Debian 11, Kernel: 5.10.0-10-amd64 (x86_64), Display Server: X Server, Vulkan: 1.0.2, Compiler: GCC 10.2.1 20210110, File-System: ext4, Screen Resolution: 1024x768 B: Processor: Intel Xeon Silver 4216 @ 3.20GHz (16 Cores / 32 Threads), Motherboard: TYAN S7100AG2NR (V4.02 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 46GB, Disk: 240GB Corsair Force MP500, Graphics: ASPEED, Audio: Realtek ALC892, Network: 2 x Intel I350 OS: Debian 11, Kernel: 5.10.0-10-amd64 (x86_64), Display Server: X Server, Vulkan: 1.0.2, Compiler: GCC 10.2.1 20210110, File-System: ext4, Screen Resolution: 1024x768 C: Processor: Intel Xeon Silver 4216 @ 3.20GHz (16 Cores / 32 Threads), Motherboard: TYAN S7100AG2NR (V4.02 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 46GB, Disk: 240GB Corsair Force MP500, Graphics: ASPEED, Audio: Realtek ALC892, Network: 2 x Intel I350 OS: Debian 11, Kernel: 5.10.0-10-amd64 (x86_64), Display Server: X Server, Vulkan: 1.0.2, Compiler: GCC 10.2.1 20210110, File-System: ext4, Screen Resolution: 1024x768 SVT-AV1 1.0 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 1.635 |=================================================================== B . 1.639 |==================================================================== C . 1.650 |==================================================================== SVT-AV1 1.0 Encoder Mode: Preset 8 - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 25.05 |==================================================================== B . 25.05 |==================================================================== C . 24.99 |==================================================================== SVT-AV1 1.0 Encoder Mode: Preset 10 - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 61.10 |=================================================================== B . 61.29 |==================================================================== C . 61.60 |==================================================================== SVT-AV1 1.0 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better A . 81.72 |==================================================================== B . 81.53 |==================================================================== C . 80.67 |=================================================================== SVT-AV1 1.0 Encoder Mode: Preset 4 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 3.850 |==================================================================== B . 3.814 |=================================================================== C . 3.808 |=================================================================== SVT-AV1 1.0 Encoder Mode: Preset 8 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 72.99 |=================================================================== B . 73.31 |==================================================================== C . 73.67 |==================================================================== SVT-AV1 1.0 Encoder Mode: Preset 10 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 141.73 |=================================================================== B . 137.80 |================================================================= C . 139.76 |================================================================== SVT-AV1 1.0 Encoder Mode: Preset 12 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better A . 197.84 |================================================================= B . 203.24 |=================================================================== C . 196.81 |================================================================= libavif avifenc 0.10 Encoder Speed: 0 Seconds < Lower Is Better A . 186.62 |=================================================================== B . 187.18 |=================================================================== C . 187.09 |=================================================================== libavif avifenc 0.10 Encoder Speed: 2 Seconds < Lower Is Better A . 90.27 |==================================================================== B . 90.41 |==================================================================== C . 90.62 |==================================================================== libavif avifenc 0.10 Encoder Speed: 6 Seconds < Lower Is Better A . 10.83 |==================================================================== B . 10.90 |==================================================================== C . 10.68 |=================================================================== libavif avifenc 0.10 Encoder Speed: 6, Lossless Seconds < Lower Is Better A . 14.76 |==================================================================== B . 14.84 |==================================================================== C . 14.71 |=================================================================== libavif avifenc 0.10 Encoder Speed: 10, Lossless Seconds < Lower Is Better A . 8.084 |=================================================================== B . 8.070 |=================================================================== C . 8.231 |==================================================================== oneDNN 2.6 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 5.32082 |================================================================== B . 4.72328 |=========================================================== C . 4.67205 |========================================================== oneDNN 2.6 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 3.90431 |================================================================ B . 3.94553 |================================================================= C . 4.01146 |================================================================== oneDNN 2.6 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 1.04894 |================================================================= B . 1.05954 |================================================================== C . 1.04970 |================================================================= oneDNN 2.6 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 1.48281 |================================================================= B . 1.50439 |================================================================== C . 1.48121 |================================================================= oneDNN 2.6 Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 11.82 |==================================================================== B . 10.64 |============================================================= C . 10.65 |============================================================= oneDNN 2.6 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 3.63975 |================================================================== B . 3.14260 |========================================================= C . 3.15259 |========================================================= oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 6.58538 |================================================================ B . 6.65136 |================================================================= C . 6.74061 |================================================================== oneDNN 2.6 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 12.69 |==================================================================== B . 11.99 |================================================================ C . 10.52 |======================================================== oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 7.80796 |================================================================ B . 7.98978 |================================================================== C . 7.80472 |================================================================ oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 6.27070 |================================================================= B . 6.35697 |================================================================== C . 6.04499 |=============================================================== oneDNN 2.6 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 1.29725 |================================================================= B . 1.31469 |================================================================== C . 1.27912 |================================================================ oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 1.88434 |================================================================== B . 1.88423 |================================================================== C . 1.88414 |================================================================== oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 3562.89 |================================================================== B . 3539.49 |================================================================== C . 3500.98 |================================================================= oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 1804.47 |=============================================================== B . 1830.47 |================================================================ C . 1885.70 |================================================================== oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 3518.89 |=============================================================== B . 3488.39 |============================================================== C . 3698.50 |================================================================== oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 16.29 |==================================================================== B . 16.29 |==================================================================== C . 16.29 |==================================================================== oneDNN 2.6 Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 21.09 |==================================================================== B . 20.94 |==================================================================== C . 20.48 |================================================================== oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 21.82 |==================================================================== B . 21.80 |==================================================================== C . 21.80 |==================================================================== oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 1819.46 |================================================================= B . 1830.52 |================================================================== C . 1839.07 |================================================================== oneDNN 2.6 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU ms < Lower Is Better A . 1.47091 |================================================================= B . 1.46191 |================================================================ C . 1.50019 |================================================================== oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 3520.29 |============================================================== B . 3718.04 |================================================================== C . 3563.23 |=============================================================== oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 1949.93 |================================================================== B . 1855.09 |=============================================================== C . 1798.18 |============================================================= oneDNN 2.6 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better A . 0.455180 |===================================================== B . 0.554475 |================================================================= C . 0.459177 |====================================================== oneDNN 2.6 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better A . 3.60266 |================================================================== B . 3.59854 |================================================================== C . 3.60331 |================================================================== ONNX Runtime 1.11 Model: GPT-2 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better A . 6555 |========================================================= B . 7962 |===================================================================== C . 6558 |========================================================= ONNX Runtime 1.11 Model: yolov4 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better A . 415 |====================================================================== B . 272 |============================================== C . 272 |============================================== ONNX Runtime 1.11 Model: bertsquad-12 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better A . 412 |============================================== B . 621 |====================================================================== C . 412 |============================================== ONNX Runtime 1.11 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better A . 45 |======================================================================= B . 45 |======================================================================= C . 45 |======================================================================= ONNX Runtime 1.11 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better A . 736 |================================================= B . 1032 |===================================================================== C . 1030 |===================================================================== ONNX Runtime 1.11 Model: super-resolution-10 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better A . 2672 |================================================================= B . 2836 |===================================================================== C . 2691 |=================================================================