Preperations

Prerequisites¶

See Prerequisites section above.

Prepare Installation Artifacts¶

Run:ai Software Files¶

SSH into a node with kubectl access to the cluster and Docker installed.

ConnectedAirgapped

Run the following to enable image download from the Run:ai Container Registry on Google cloud:

kubectl create namespace runai-backend
kubectl apply -f runai-gcr-secret.yaml

To extract Run:ai files, replace <VERSION> in the command below and run:

tar xvf runai-<version>.tar.gz
cd deploy

Upload images to Docker Registry. Set the Docker Registry address in the form of NAME:PORT (do not add https):

export REGISTRY_URL=<Docker Registry address>

Run the following script (you must have at least 20GB of free disk space to run):

kubectl create namespace runai-backend

sudo -E ./prepare_installation.sh

(If docker is configured to run as non-root then sudo is not required).

Run:ai Administration CLI¶

ConnectedAirgapped

Install the Run:ai Administrator Command-line Interface by following the steps here.

Install the Run:ai Administrator Command-line Interface by following the steps here. Use the image under deploy/runai-admin-cli-<version>-linux-amd64.tar.gz

Install Helm¶

If helm v3 does not yet exist on the machine, install it now:

ConnectedAirgapped

See https://helm.sh/docs/intro/install/ on how to install Helm. Run:ai works with Helm version 3 only (not helm 2).

The Helm installation image is under the deploy directory. Run:

tar xvf helm-<version>-linux-amd64.tar.gz
sudo mv linux-amd64/helm /usr/local/bin/

Mark Run:ai System Workers¶

The Run:ai control plane (backend) should be installed on a set of dedicated Run:ai system worker nodes rather than GPU worker nodes. To set system worker nodes run:

kubectl label node <NODE-NAME> node-role.kubernetes.io/runai-system=true

To avoid single-point-of-failure issues, we recommend assigning more than one node in production environments.

Additional Permissions¶

As part of the installation you will be required to install the Run:ai Control Plane and Cluster Helm Charts. The Helm Charts require Kubernetes administrator permissions. You can review the exact permissions provided by using the --dry-run on both helm charts.

Next Steps¶

Continue with installing the Run:ai Control Plane.

Last update: June 19, 2022