Hi Community,
Just as the title says, any know why MIG instances can only sum up to 98 SMs instead of the 108 SMs available in the A100 Tensor Core GPU Implementation (https://developer.nvidia.com/blog/nvidia-ampere-architecture-in-depth/)? Any explanation is welcome. Best
➜ ~ sudo nvidia-smi mig -lgip
+--------------------------------------------------------------------------+
| GPU instance profiles: |
| GPU Name ID Instances Memory P2P SM DEC ENC |
| Free/Total GiB CE JPEG OFA |
|==========================================================================|
| 4 MIG 1g.5gb 19 0/7 4.75 No 14 0 0 |
| 1 0 0 |
+--------------------------------------------------------------------------+
| 4 MIG 2g.10gb 14 0/3 9.75 No 28 1 0 |
| 2 0 0 |
+--------------------------------------------------------------------------+
| 4 MIG 3g.20gb 9 0/2 19.62 No 42 2 0 |
| 3 0 0 |
+--------------------------------------------------------------------------+
| 4 MIG 4g.20gb 5 0/1 19.62 No 56 2 0 |
| 4 0 0 |
+--------------------------------------------------------------------------+
| 4 MIG 7g.40gb 0 0/1 39.50 No 98 5 0 |
| 7 1 1 |
+--------------------------------------------------------------------------+