This is my attempt to make the GPU monitoring work on p-type instances. p3 instances are indeed recognized and show up in the GPU Nodes list after applying this patch. Also, the nvidia-dcgm monitoring seems to be running on the nodes.
However, I still don't see any GPU data in Grafana. Everything is showing "no data" or "N/A".
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.
This is my attempt to make the GPU monitoring work on p-type instances. p3 instances are indeed recognized and show up in the GPU Nodes list after applying this patch. Also, the nvidia-dcgm monitoring seems to be running on the nodes.
However, I still don't see any GPU data in Grafana. Everything is showing "no data" or "N/A".
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.