I’m running 4 80GB A100 SXM4’s, it appears each card is limited to only 275w each. However, I’ve seen that the TDP should be closer to 400w. I have no way to increase it, but I believe it’s limiting my processing power for the jobs I receive. Does anybody have any idea why that would be happening? It has a 2000w PSU so I don’t believe the PSU is the problem.
±--------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.183.06 Driver Version: 535.183.06 CUDA Version: 12.2 |
|-----------------------------------------±---------------------±---------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 NVIDIA A100-SXM4-80GB On | 00000000:01:00.0 Off | Off |
| N/A 41C P0 60W / 275W | 0MiB / 81920MiB | 0% Default |
| | | Disabled |
±----------------------------------------±---------------------±---------------------+
| 1 NVIDIA A100-SXM4-80GB On | 00000000:47:00.0 Off | Off |
| N/A 41C P0 65W / 275W | 0MiB / 81920MiB | 0% Default |
| | | Disabled |
±----------------------------------------±---------------------±---------------------+
| 2 NVIDIA A100-SXM4-80GB On | 00000000:81:00.0 Off | Off |
| N/A 40C P0 62W / 275W | 0MiB / 81920MiB | 0% Default |
| | | Disabled |
±----------------------------------------±---------------------±---------------------+
| 3 NVIDIA A100-SXM4-80GB On | 00000000:C2:00.0 Off | Off |
| N/A 40C P0 61W / 275W | 0MiB / 81920MiB | 0% Default |
| | | Disabled |
±----------------------------------------±---------------------±---------------------+
±--------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| No running processes found |
±--------------------------------------------------------------------------------------+