Get Started with Grafana

Trainy: Konduktor users can connect their telemetry to Grafana, featuring pre-configured dashboards optimized to display the metrics, events, and logs most relevant to your workloads.

To access your Grafana instance, contact Trainy support for access to the Grafana URL and login credentials.

Grafana Explore

Grafana Explore enables users to do full PromQL and LogQL queries over logs and metrics and filter on metadata such as workload title or username.

Trainy provides default dashboards curated views of logs and metrics, while the Explore tab empowers users to drill down into telemetry for in depth debugging.

Available Dashboards

Cluster Overview

The cluster overview dashboard shows overall system utilization such as GPU utilization, efficiency, Infiniband/RoCE throughput, and NVLINK throughput.

Cluster view

Workload Inspector

The workload inspector (video) allows for filtering of system metrics by workload and username.

user filter

workload filter