How to monitor multiple GPU servers

NVIDIA / gpu-monitoring-tools

Tools for monitoring NVIDIA GPUs on Linux

Apache License 2.0

1.02k stars 301 forks source link

How to monitor multiple GPU servers #181

Open anilnokia opened 3 years ago

anilnokia commented 3 years ago

Hi,

I installed Nvidia DGCM exporter and tools and got the GPU dashboard into grafana. It works well for one server. But I have multiple GPU servers and even if I add into prometheus, I only see one server stats in to grafana. There is no dropdown to select servers. Is this possible? How can I see diffrent servers.

Thanks,

Anil Rana Nokia

nikkon-dev commented 3 years ago

Hi,

The latest versions of the dcgm-exporter should output hostname as a label in each metric until the --no-hostname or a related ENV variable is set.

Would that help you to set up your Grafana dashboards?

WBR, Nik