Skip to content

NVIDIA Device Plugin

The NVIDIA device plugin for Kubernetes is a Daemonset that allows you to automatically:

  • Expose the number of GPUs on each nodes of your cluster
  • Keep track of the health of your GPUs
  • Run GPU enabled containers in your Kubernetes cluster.

For complete project documentation, please visit the NVIDIA Device Plugin.

Additionally, refer to this AWS blog for more information on how the add-on can be tested.

Usage

NVIDIA device plugin can be deployed by enabling the add-on via the following.

enable_nvidia_device_plugin = true

You can optionally customize the Helm chart via the following configuration.

  enable_nvidia_device_plugin = true
  # Optional nvidia_device_plugin_helm_config
  nvidia_device_plugin_helm_config = {
    name                       = "nvidia-device-plugin"
    chart                      = "nvidia-device-plugin"
    repository                 = "https://nvidia.github.io/k8s-device-plugin"
    version                    = "0.12.3"
    namespace                  = "nvidia-device-plugin"
    values = [templatefile("${path.module}/values.yaml", {
      ...
    })]
  }

GitOps Configuration

The following properties are made available for use when managing the add-on via GitOps.

Refer to locals.tf for latest config. GitOps with ArgoCD Add-on repo is located here

  argocd_gitops_config = {
    enable = true
  }