Enabling Quota Management in Cloudera Machine Learning

To enable Quota Management in Cloudera Machine Learning it needs to be configured. Follow the recommended configuration guidelines.

Set up the Kubernetes and kubectl as defined in Prerequisites.

  1. Configure the kubectl using the kubeconfig file.
  2. Edit the Cloudera Machine Learning control plane deployment:
    kubectl get deploy dp-mlx-control-plane-app -n cdp-namespace -o yaml
    > file-name
    This will save the Cloudera Machine Learning control plane deployment specification.
  3. Edit the Cadence Worker deployment with the following command:
    kubectl get deploy dp-cadence-worker -n <cdp-namespace> -o yaml > <cadence-file-name>

    The command saves Cadence Worker deployment specification.

  4. Take a backup of the above files.
  5. Search for the environment variable ENABLE_UMBRA_INTEGRATION in both files, and change the value to true.
  6. Save the deployment files and run the following commands:
    kubectl apply -f file-name.yaml
    kubectl apply -f <cadence-file-name>.yaml
    Wait for the new pods to come up.
  7. Verify the new pods by running the following command:
    kubectl get pods -n cdp-namespace