mpi-operator

packaged by Chainguard

Chainguard Container for mpi-operator

Minimal MPI Operator image for managing distributed MPI workloads on Kubernetes.

Chainguard Containers are regularly-updated, secure-by-default container images.

Download this Container Image

For those with access, this container image is available on cgr.dev:

docker pull cgr.dev/ORGANIZATION/mpi-operator:latest

Be sure to replace the ORGANIZATION placeholder with the name used for your organization's private repository within the Chainguard Registry.

Compatibility Notes

The cgr.dev/ORGANIZATION/mpi-operator image is comparable to the upstream mpioperator/mpi-operator image. It preserves the /opt/mpi-operator entrypoint and the default --help argument. The controller runs as the nonroot user with UID 65532 instead of upstream's root user, reducing the privileges available to the process. The upstream standalone deployment supports this configuration.

This image contains the Kubernetes controller. MPI launcher and worker images, including the sample Pi workload, are separate images and must provide the MPI implementation required by your jobs.

Prerequisites

You need a Kubernetes cluster, kubectl with Kustomize support, Git on your PATH, and permission to install custom resource definitions and cluster RBAC. Kustomize uses Git to load the upstream manifests. Configure your cluster to pull the image from your organization's registry repository. See authenticating to Chainguard's registry.

Getting Started

Deploy the controller with the upstream standalone Kustomize overlay. Replace x.y.z with the image's release version, without a leading v, and ORGANIZATION with your organization's registry repository name. The same version selects the controller image and its matching upstream manifests:

MPI_OPERATOR_VERSION=x.y.z
mkdir -p mpi-operator
cat > mpi-operator/kustomization.yaml <<EOF
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization
resources:
  - https://github.com/kubeflow/mpi-operator/manifests/overlays/standalone?ref=v${MPI_OPERATOR_VERSION}
images:
  - name: mpioperator/mpi-operator
    newName: cgr.dev/ORGANIZATION/mpi-operator
    newTag: ${MPI_OPERATOR_VERSION}
EOF
kubectl apply --server-side -k mpi-operator
kubectl rollout status deployment/mpi-operator -n mpi-operator --timeout=180s

The overlay installs the MPIJob custom resource definition, service account, RBAC, and controller deployment in the mpi-operator namespace. See the Pi example in the upstream repository's examples folder to submit an MPIJob with launcher and worker images suited to your MPI runtime.

Pass --hostfile /etc/mpi/hostfile in the MPIJob launcher's mpirun arguments to explicitly select the worker hostfile generated by the controller. This ensures the workload runs on the worker Pods rather than relying on the launcher's default host selection.

Configuration

The controller monitors MPIJobs across the cluster by default. To monitor a single workload namespace, append a deployment patch to the overlay before applying it. This example creates the mpi-jobs namespace and adds the controller's --namespace argument while preserving the upstream arguments:

kubectl create namespace mpi-jobs
cat >> mpi-operator/kustomization.yaml <<'EOF'
patches:
  - target:
      group: apps
      version: v1
      kind: Deployment
      name: mpi-operator
    patch: |-
      - op: add
        path: /spec/template/spec/containers/0/args/-
        value: --namespace=mpi-jobs
EOF
kubectl apply --server-side -k mpi-operator
kubectl rollout status deployment/mpi-operator -n mpi-operator --timeout=180s

Submit your MPIJobs in mpi-jobs when using this configuration. The controller's leader election resources remain in the mpi-operator namespace.

Documentation and Resources

• MPI Operator documentation
• MPI Operator releases
• MPIJob examples
• Chainguard Containers documentation

What are Chainguard Containers?

Chainguard's free tier of Starter container images are built with Wolfi, our minimal Linux undistro.

All other Chainguard Containers are built with Chainguard OS, Chainguard's minimal Linux operating system designed to produce container images that meet the requirements of a more secure software supply chain.

The main features of Chainguard Containers include:

• Minimal design, without unnecessary software bloat
• Daily builds to ensure container images are up-to-date with available security patches
• High quality build-time SBOMs attesting to the provenance of all artifacts within the image
• Verifiable signatures provided by Sigstore
• Reproducible builds with Cosign and apko (read more about reproducibility)

For cases where you need container images with shells and package managers to build or debug, most Chainguard Containers come paired with a development, or -dev, variant.

In all other cases, including Chainguard Containers tagged as :latest or with a specific version number, the container images include only an open-source application and its runtime dependencies. These minimal container images typically do not contain a shell or package manager.

Although the -dev container image variants have similar security features as their more minimal versions, they include additional software that is typically not necessary in production environments. We recommend using multi-stage builds to copy artifacts from the -dev variant into a more minimal production image.

Need additional packages?

To improve security, Chainguard Containers include only essential dependencies. Need more packages? Chainguard customers can use Custom Assembly to add packages, either through the Console, chainctl, or API.

To use Custom Assembly in the Chainguard Console: navigate to the image you'd like to customize in your Organization's list of images, and click on the Customize image button at the top of the page.

Learn More

Refer to our Chainguard Containers documentation on Chainguard Academy. Chainguard also offers VMs and Libraries — contact us for access.

Trademarks

This software listing is packaged by Chainguard. The trademarks set forth in this offering are owned by their respective companies, and use of them does not imply any affiliation, sponsorship, or endorsement by such companies.