GPU usage when pod initialized. Not able to clear.
Tried nvidia-smi -r, restarting, and reseting. There is still usage on one gpu in the pod.
4 Replies
@zkreutzjanz I've noted the pod/machine ID so we can take a look internally, best thing to do would be to rent another set of GPUs in the meantime if one is available. Sorry for the incovenience
@Finley FYI I am still renting this pod. Can't take it down due to needing a bunch of A6000s on the same machine. The extra memory usage is still there.