Installation#

Choose the guide for the component you want to install. Install DCGM to use its host engine, command-line tools, diagnostics, or APIs. Install DCGM Exporter to expose selected DCGM telemetry to Prometheus.

Choose What to Install#

Installation guides#

Component

Includes

Use it when

Guide

DCGM

The DCGM host engine, dcgmi, libraries, diagnostics, field definitions, and development files.

A host needs DCGM management, health, diagnostics, telemetry, or API access.

Install DCGM

DCGM Exporter

A Prometheus /metrics endpoint backed by selected DCGM fields.

Prometheus needs GPU, NVSwitch, NVIDIA CPU, or workload-attributed telemetry from a host or Kubernetes cluster.

Install DCGM Exporter

How the Components Relate#

DCGM Exporter uses DCGM through libdcgm. By default, it initializes an embedded DCGM host engine inside the exporter process. It can instead connect to a separately managed DCGM host engine over a supported remote transport.

The installation method determines who installs and manages DCGM:

  • With packages, the system package manager updates DCGM and DCGM Exporter, and systemd manages their services. Install the tested DCGM and exporter versions together when you upgrade.

  • A standalone exporter container carries the DCGM libraries selected for its release. Its image tag selects the exporter version; the release notes list the tested DCGM version. It can use embedded or remote operation.

  • A Helm release manages the exporter DaemonSet and related Kubernetes resources. Its chart values select the exporter image and tag; the release notes list the tested DCGM version.

  • NVIDIA GPU Operator manages DCGM, DCGM Exporter, and the surrounding GPU software stack as one stack.

Follow the selected guide. Not every exporter deployment needs a separately installed DCGM host engine.

Shared Prerequisites#

Both products require a supported NVIDIA GPU, driver, operating environment, and access to the GPU devices. Package installation and service changes require administrative privileges. Container deployments also require the NVIDIA Container Toolkit and permission to use the container runtime.

Use the DCGM product support matrix and the detailed requirements in the selected installation guide before deploying across a fleet.

Next Steps#