How to get pod level GPU metrics

NVIDIA / gpu-monitoring-tools

Tools for monitoring NVIDIA GPUs on Linux

Apache License 2.0

1.02k stars 301 forks source link

How to get pod level GPU metrics #214

Open faheemsohail opened 3 years ago

faheemsohail commented 3 years ago

I have setup dcgm-exporter to collect metrics for GPU usage of pods but the pod field shows the name of dcgm-exporter and not the actual pod generating the workload.

pod="dcgm-exporter-1634736248-7c6vs"

Is there a config to be made in order to get pod level GPU metrics?