GPU.7XLP100 vs NVIDIA DGX-1
Geekbench Scores
| GPU.7XLP100 | NVIDIA DGX-1 | |||
|---|---|---|---|---|
| CUDA Score | 320031 | 743537 | ||
| Geekbench 4.0.4 | Geekbench 4.1.1 | |||
System Information
| GPU.7XLP100 | NVIDIA DGX-1 | |
|---|---|---|
| Operating System | Ubuntu 16.04.2 LTS 4.4.0-64-generic x86_64 | Ubuntu 16.04.2 LTS 4.4.0-62-generic x86_64 |
| Model | VMware, Inc. VMware7,1 | NVIDIA DGX-1 |
| Processor |
Intel Xeon E5-2690 v4 @ 2.60 GHz
56 Processors, 56 Cores |
Intel Xeon E5-2698 v4 @ 3.60 GHz
2 Processors, 40 Cores, 80 Threads |
| Processor ID | GenuineIntel Family 6 Model 79 Stepping 1 | GenuineIntel Family 6 Model 79 Stepping 1 |
| L1 Instruction Cache | 32.0 KB x 1 | 32.0 KB x 20 |
| L1 Data Cache | 32.0 KB x 1 | 32.0 KB x 20 |
| L2 Cache | 256 KB x 1 | 256 KB x 20 |
| L3 Cache | 35.0 MB x 1 | 50.0 MB x 1 |
| Motherboard | Intel Corporation 440BX Desktop Reference Platform | NVIDIA DGX-1 |
| BIOS | VMware, Inc. VMW71.00V.0.B64.1506250318 | American Megatrends Inc. S2W_3A03 |
| Memory | 251.90 GB | 503.80 GB |
| CUDA Information | ||
| Device Name | Tesla P100-PCIE-16GB | Tesla V100-SXM2-16GB |
| Compute Capability Memory | 6.0 | 7.0 |
| Maximum Frequency | 1.33 GHz | 1.80 GHz |
| Multiprocessor Count | 56 | 80 |
| Maximum Threads Per Multiprocessor | 2048 | 2048 |
| Device Memory | 15.9 GB 715 MHz | 15.8 GB 877 MHz |
CUDA Performance
| GPU.7XLP100 | NVIDIA DGX-1 | |||
|---|---|---|---|---|
| CUDA Score | 320031 | 743537 | ||
| Sobel |
528482
23.3 Gpixels/sec |
1382119
60.9 Gpixels/sec |
||
| Histogram Equalization |
455379
14.2 Gpixels/sec |
996475
31.1 Gpixels/sec |
||
| SFFT |
66489
165.7 Gflops |
101670
253.5 Gflops |
||
| Gaussian Blur |
538403
9.43 Gpixels/sec |
1897300
33.2 Gpixels/sec |
||
| Face Detection |
49263
14.4 Msubwindows/sec |
108700
31.7 Msubwindows/sec |
||
| RAW |
1139825
11.0 Gpixels/sec |
2743361
26.6 Gpixels/sec |
||
| Depth of Field |
571644
1.66 Gpixels/sec |
1499040
4.35 Gpixels/sec |
||
| Particle Physics |
397917
62904.7 FPS |
786603
124350.1 FPS |
||
