Intel Xeon Silver 4216 testing with a TYAN S7100AG2NR (V4.02 BIOS) and ASPEED on Debian 12 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2401144-NE-XEONJAN1706
xeon jan
Intel Xeon Silver 4216 testing with a TYAN S7100AG2NR (V4.02 BIOS) and ASPEED on Debian 12 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: 6 x 8 GB DDR4-2400MT/s, Disk: 240GB Corsair Force MP500, Graphics: ASPEED, Audio: Realtek ALC892, Network: 2 x Intel I350
OS: Debian 12, Kernel: 6.1.0-11-amd64 (x86_64), Display Server: X Server, Compiler: GCC 12.2.0, 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: 6 x 8 GB DDR4-2400MT/s, Disk: 240GB Corsair Force MP500, Graphics: ASPEED, Audio: Realtek ALC892, Network: 2 x Intel I350
OS: Debian 12, Kernel: 6.1.0-11-amd64 (x86_64), Display Server: X Server, Compiler: GCC 12.2.0, 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: 6 x 8 GB DDR4-2400MT/s, Disk: 240GB Corsair Force MP500, Graphics: ASPEED, Audio: Realtek ALC892, Network: 2 x Intel I350
OS: Debian 12, Kernel: 6.1.0-11-amd64 (x86_64), Display Server: X Server, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1024x768
Quicksilver 20230818
Input: CTS2
Figure Of Merit > Higher Is Better
a . 8446000 |=================================================================
b . 8497000 |=================================================================
c . 8607000 |==================================================================
Quicksilver 20230818
Input: CORAL2 P2
Figure Of Merit > Higher Is Better
a . 9287000 |==================================================================
b . 9354000 |==================================================================
c . 9308000 |==================================================================
PyTorch 2.1
Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
a . 4.72 |====================================================================
b . 4.76 |=====================================================================
c . 4.70 |====================================================================
PyTorch 2.1
Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
a . 4.79 |=====================================================================
b . 4.75 |====================================================================
c . 4.75 |====================================================================
LeelaChessZero 0.30
Backend: BLAS
Nodes Per Second > Higher Is Better
a . 37 |=====================================================================
b . 38 |=======================================================================
c . 37 |=====================================================================
LeelaChessZero 0.30
Backend: Eigen
Nodes Per Second > Higher Is Better
a . 33 |=======================================================================
b . 33 |=======================================================================
c . 32 |=====================================================================
Llama.cpp b1808
Model: llama-2-70b-chat.Q5_0.gguf
Tokens Per Second > Higher Is Better
a . 1.50 |=====================================================================
b . 1.51 |=====================================================================
c . 1.50 |=====================================================================
TensorFlow 2.12
Device: CPU - Batch Size: 16 - Model: VGG-16
images/sec > Higher Is Better
a . 5.96 |=====================================================================
b . 5.96 |=====================================================================
c . 5.96 |=====================================================================
PyTorch 2.1
Device: CPU - Batch Size: 32 - Model: ResNet-152
batches/sec > Higher Is Better
a . 8.18 |=====================================================================
b . 8.09 |====================================================================
c . 8.08 |====================================================================
PyTorch 2.1
Device: CPU - Batch Size: 16 - Model: ResNet-152
batches/sec > Higher Is Better
a . 8.14 |====================================================================
b . 8.13 |====================================================================
c . 8.22 |=====================================================================
PyTorch 2.1
Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
a . 6.91 |=====================================================================
b . 6.90 |=====================================================================
c . 6.95 |=====================================================================
CacheBench
Test: Read / Modify / Write
MB/s > Higher Is Better
a . 61680.56 |=================================================================
b . 59877.34 |===============================================================
c . 60843.70 |================================================================
CacheBench
Test: Write
MB/s > Higher Is Better
a . 23161.61 |=================================================================
b . 23134.97 |=================================================================
c . 23165.59 |=================================================================
CacheBench
Test: Read
MB/s > Higher Is Better
a . 6062.37 |==================================================================
b . 6058.74 |==================================================================
c . 6057.99 |==================================================================
TensorFlow 2.12
Device: CPU - Batch Size: 16 - Model: ResNet-50
images/sec > Higher Is Better
a . 16.22 |====================================================================
b . 16.25 |====================================================================
c . 16.21 |====================================================================
Quicksilver 20230818
Input: CORAL2 P1
Figure Of Merit > Higher Is Better
a . 10170000 |=================================================================
b . 10110000 |=================================================================
c . 10150000 |=================================================================
PyTorch 2.1
Device: CPU - Batch Size: 1 - Model: ResNet-152
batches/sec > Higher Is Better
a . 11.11 |====================================================================
b . 11.17 |====================================================================
c . 11.07 |===================================================================
Neural Magic DeepSparse 1.6
Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 845.95 |===================================================================
b . 846.00 |===================================================================
c . 845.20 |===================================================================
Neural Magic DeepSparse 1.6
Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 9.4169 |===================================================================
b . 9.4557 |===================================================================
c . 9.4646 |===================================================================
PyTorch 2.1
Device: CPU - Batch Size: 16 - Model: ResNet-50
batches/sec > Higher Is Better
a . 21.57 |===================================================================
b . 21.27 |===================================================================
c . 21.73 |====================================================================
PyTorch 2.1
Device: CPU - Batch Size: 32 - Model: ResNet-50
batches/sec > Higher Is Better
a . 21.56 |===================================================================
b . 21.64 |====================================================================
c . 21.74 |====================================================================
Llama.cpp b1808
Model: llama-2-13b.Q4_0.gguf
Tokens Per Second > Higher Is Better
a . 8.70 |=====================================================================
b . 8.73 |=====================================================================
c . 8.62 |====================================================================
SVT-AV1 1.8
Encoder Mode: Preset 4 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
a . 2.462 |====================================================================
b . 2.423 |===================================================================
c . 2.421 |===================================================================
Speedb 2.7
Test: Random Fill Sync
Op/s > Higher Is Better
a . 8962 |=============================================
b . 13397 |====================================================================
c . 10150 |====================================================
Speedb 2.7
Test: Random Fill
Op/s > Higher Is Better
a . 379730 |===================================================================
b . 298026 |=====================================================
c . 377206 |===================================================================
Speedb 2.7
Test: Update Random
Op/s > Higher Is Better
a . 172891 |===================================================================
b . 163726 |===============================================================
c . 151137 |===========================================================
Speedb 2.7
Test: Read While Writing
Op/s > Higher Is Better
a . 3897119 |================================================================
b . 3867484 |================================================================
c . 4014397 |==================================================================
Speedb 2.7
Test: Read Random Write Random
Op/s > Higher Is Better
a . 1640953 |=================================================================
b . 1658156 |==================================================================
c . 1656172 |==================================================================
Speedb 2.7
Test: Random Read
Op/s > Higher Is Better
a . 53271554 |=================================================================
b . 52915603 |=================================================================
c . 52443533 |================================================================
Speedb 2.7
Test: Sequential Fill
Op/s > Higher Is Better
a . 565169 |===================================================================
b . 558662 |==================================================================
c . 549382 |=================================================================
Neural Magic DeepSparse 1.6
Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 28.14 |====================================================================
b . 27.85 |===================================================================
c . 27.95 |====================================================================
Neural Magic DeepSparse 1.6
Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 283.96 |==================================================================
b . 286.91 |===================================================================
c . 285.94 |===================================================================
Neural Magic DeepSparse 1.6
Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 1061.89 |=================================================================
b . 1073.13 |==================================================================
c . 1060.60 |=================================================================
Neural Magic DeepSparse 1.6
Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 7.5161 |===================================================================
b . 7.2935 |=================================================================
c . 7.5416 |===================================================================
Neural Magic DeepSparse 1.6
Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 1071.41 |==================================================================
b . 1063.84 |=================================================================
c . 1072.98 |==================================================================
Neural Magic DeepSparse 1.6
Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 7.3201 |=================================================================
b . 7.5195 |===================================================================
c . 7.2847 |=================================================================
Y-Cruncher 0.8.3
Pi Digits To Calculate: 1B
Seconds < Lower Is Better
a . 46.09 |====================================================================
b . 45.45 |===================================================================
c . 45.93 |====================================================================
Neural Magic DeepSparse 1.6
Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 63.48 |====================================================================
b . 63.39 |====================================================================
c . 63.53 |====================================================================
Neural Magic DeepSparse 1.6
Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 125.96 |===================================================================
b . 126.10 |===================================================================
c . 125.77 |===================================================================
Neural Magic DeepSparse 1.6
Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 552.53 |===================================================================
b . 549.40 |===================================================================
c . 553.47 |===================================================================
Neural Magic DeepSparse 1.6
Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 14.38 |===================================================================
b . 14.50 |====================================================================
c . 14.30 |===================================================================
PyTorch 2.1
Device: CPU - Batch Size: 1 - Model: ResNet-50
batches/sec > Higher Is Better
a . 29.51 |====================================================================
b . 29.64 |====================================================================
c . 28.89 |==================================================================
Neural Magic DeepSparse 1.6
Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 123.22 |===================================================================
b . 123.89 |===================================================================
c . 123.78 |===================================================================
Neural Magic DeepSparse 1.6
Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 64.91 |====================================================================
b . 64.54 |====================================================================
c . 64.62 |====================================================================
TensorFlow 2.12
Device: CPU - Batch Size: 16 - Model: GoogLeNet
images/sec > Higher Is Better
a . 47.63 |====================================================================
b . 47.51 |====================================================================
c . 47.36 |====================================================================
Neural Magic DeepSparse 1.6
Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 68.77 |====================================================================
b . 68.91 |====================================================================
c . 68.89 |====================================================================
Neural Magic DeepSparse 1.6
Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 116.28 |===================================================================
b . 116.04 |===================================================================
c . 116.08 |===================================================================
Neural Magic DeepSparse 1.6
Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 164.63 |===================================================================
b . 164.59 |===================================================================
c . 164.08 |===================================================================
Neural Magic DeepSparse 1.6
Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 48.57 |====================================================================
b . 48.60 |====================================================================
c . 48.65 |====================================================================
Neural Magic DeepSparse 1.6
Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 165.97 |===================================================================
b . 165.69 |===================================================================
c . 165.88 |===================================================================
Neural Magic DeepSparse 1.6
Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 48.19 |====================================================================
b . 48.27 |====================================================================
c . 48.22 |====================================================================
Neural Magic DeepSparse 1.6
Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 10.90 |====================================================================
b . 10.84 |====================================================================
c . 10.87 |====================================================================
Neural Magic DeepSparse 1.6
Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 732.36 |===================================================================
b . 736.87 |===================================================================
c . 734.42 |===================================================================
Neural Magic DeepSparse 1.6
Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
a . 68.76 |====================================================================
b . 68.95 |====================================================================
c . 68.68 |====================================================================
Neural Magic DeepSparse 1.6
Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
a . 116.18 |===================================================================
b . 115.88 |===================================================================
c . 116.40 |===================================================================
TensorFlow 2.12
Device: CPU - Batch Size: 1 - Model: VGG-16
images/sec > Higher Is Better
a . 3.27 |=====================================================================
b . 3.24 |====================================================================
c . 3.26 |=====================================================================
TensorFlow 2.12
Device: CPU - Batch Size: 1 - Model: ResNet-50
images/sec > Higher Is Better
a . 4.81 |====================================================================
b . 4.87 |=====================================================================
c . 4.88 |=====================================================================
SVT-AV1 1.8
Encoder Mode: Preset 8 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
a . 24.38 |====================================================================
b . 23.99 |===================================================================
c . 24.07 |===================================================================
TensorFlow 2.12
Device: CPU - Batch Size: 16 - Model: AlexNet
images/sec > Higher Is Better
a . 83.17 |====================================================================
b . 82.83 |====================================================================
c . 83.33 |====================================================================
SVT-AV1 1.8
Encoder Mode: Preset 4 - Input: Bosphorus 1080p
Frames Per Second > Higher Is Better
a . 7.326 |====================================================================
b . 7.379 |====================================================================
c . 7.251 |===================================================================
Y-Cruncher 0.8.3
Pi Digits To Calculate: 500M
Seconds < Lower Is Better
a . 20.62 |====================================================================
b . 20.68 |====================================================================
c . 20.58 |====================================================================
Llama.cpp b1808
Model: llama-2-7b.Q4_0.gguf
Tokens Per Second > Higher Is Better
a . 16.95 |====================================================================
b . 15.89 |================================================================
c . 16.55 |==================================================================
SVT-AV1 1.8
Encoder Mode: Preset 8 - Input: Bosphorus 1080p
Frames Per Second > Higher Is Better
a . 45.63 |==================================================================
b . 46.69 |====================================================================
c . 45.94 |===================================================================
TensorFlow 2.12
Device: CPU - Batch Size: 1 - Model: GoogLeNet
images/sec > Higher Is Better
a . 17.26 |====================================================================
b . 15.86 |==============================================================
c . 16.19 |================================================================
SVT-AV1 1.8
Encoder Mode: Preset 12 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
a . 82.81 |====================================================================
b . 78.54 |================================================================
c . 82.22 |====================================================================
SVT-AV1 1.8
Encoder Mode: Preset 13 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
a . 82.39 |===================================================================
b . 82.62 |===================================================================
c . 83.27 |====================================================================
TensorFlow 2.12
Device: CPU - Batch Size: 1 - Model: AlexNet
images/sec > Higher Is Better
a . 18.21 |===================================================================
b . 18.25 |====================================================================
c . 18.35 |====================================================================
SVT-AV1 1.8
Encoder Mode: Preset 12 - Input: Bosphorus 1080p
Frames Per Second > Higher Is Better
a . 165.27 |=================================================================
b . 170.98 |===================================================================
c . 168.01 |==================================================================
SVT-AV1 1.8
Encoder Mode: Preset 13 - Input: Bosphorus 1080p
Frames Per Second > Higher Is Better
a . 188.99 |===================================================================
b . 184.72 |=================================================================
c . 187.02 |==================================================================