Cloudera Docs

Enabling Model Governance

Cloudera Data Science Workbench utilizes Atlas lineage to help you understand the source and impact of data and changes to data over time and across all your data.

You must install the following services on your CDP cluster:
  • Atlas
  • Ranger
  • Ranger KMS
  • Kafka
  • ZooKeeper
  • SOLR
You must enable governance to capture and view information about your CDSW projects, models, and builds centrally from Apache Atlas (Data Catalog) for a given environment. If you do not select this option while provisioning workspaces, then integration with Atlas will not work.
  1. Go to Cloudera Manager.
  2. Choose your CDSW cluster.
  3. Click the Configuration tab.
  4. Check the Enable Model Governance Support checkbox for your CDSW cluster.
    Wait for Cloudera Manager to detect and validate your change. Your change will result in a Stale Configuration and Cloudera Manager will require that you restart your cluster.
  5. If you want to review your changes:
    1. Click the Restart button.
      Cloudera Manager displays a list of your changes.
    2. Click the Restart Stale Services button at the bottom of the screen.
    3. Confirm your changes by clicking the checkbox next to Re-deploy client configuration.
  6. If you do not want to review your changes, choose Restart from the Actions menu.