NVIDIA/dcgm-exporter

Health v2 · maintenance only

Adoption not part of the score

1844 stars · 326 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

NVIDIA's GPU metrics exporter for Prometheus, built on top of DCGM (Data Center GPU Manager). It exposes GPU telemetry such as clock speeds, temperature, memory usage, and utilization over an HTTP /metrics endpoint, with first-class Kubernetes deployment via Helm.

Use cases

• monitor nvidia gpu utilization and temperature in prometheus
• export gpu metrics from kubernetes nodes
• scrape dcgm gpu telemetry for grafana dashboards
• track gpu memory usage and clock speeds in a cluster
• set up gpu health monitoring with the nvidia gpu operator
• collect per-gpu metrics on a linux machine with docker

When to choose

• you run nvidia gpus on linux and want prometheus-compatible metrics
• you need gpu telemetry in kubernetes, ideally alongside the gpu operator
• you want an official, maintained exporter from nvidia

When to avoid

• you use amd or intel gpus instead of nvidia
• you need a full gpu monitoring dashboard rather than a metrics exporter
• you are not using prometheus or a prometheus-compatible scraper

Facets

service · maturity active

monitoring monitoring gpu-computing go prometheus nvidia gpu-metrics dcgm telemetry helm-chart metrics-exporter devops containers linux docker kubernetes

Member repositories

For agents

markdown · JSON · MCP: product_card(name="NVIDIA/dcgm-exporter")