Hello everyone, 1) We have kubernetes cluster it contains 6 nodes and out of 3 are GPU one, we running the LLM models on it. So have installed the signoz in our kubernetes cluster with helm chart. But not sure how can i do the live monitoring of GPU. Any have any idea on this, that will be very helpful for me 🙂

How do you generally collect the GPU monitoring data?

we don't have any mechanism on place yet, through signoz we are looking for it first time.

SigNoz will work as a backend. You need some way to collect and send them to SigNoz.

We are using a azure AKS cluster, now we are monitoring GPU through the azure portal. But we want to use signoz as central dashboard to monitor everything, so let us know how i can send and catch the gpu stats to signoz. @Ankit Nayan @Srikanth Chekuri Please help me 🙂

Have you tried using https://github.com/NVIDIA/dcgm-exporter?

it worked, thanks