Cloudera Docs

Provisioning a Cloudera AI Workbench

In Cloudera AI on on premises, the Cloudera AI Workbench provides a space for the data scientists' work. After your Administrator has created or given you access to an environment, you can set up a workbench.

  • The first user to access the Cloudera AI Workbench after it is created must have the EnvironmentAdmin role assigned.

  • If you have the cert-manager setup in your cluster, the configuration to enable the Certification Manager has to be completed prior to provisioning a Cloudera AI Workbench. For details, see Configuring cluster issuer for cert-manager. These steps are required only if you have the cert-manager setup.

    During the workspace installation process, Cloudera AI validates the cluster issuer configurations. If a cluster issuer is properly labeled and annotated, it will be utilized to sign certificates for both the workbench and workload domains.

  • The Certification Manager has a limitation that restricts common names to a maximum of 64 characters. To address this, the static subdomain feature must be used to configure a shorter Cloudera AI subdomain. During the workbench provisioning workflow, preflight checks are conducted to ensure that the Cloudera AI Workbench domain name complies with this character limit.

  1. Log in to the Cloudera on premises web interface using your corporate credentials or other credentials that you received from your Cloudera administrator.
  2. Click Cloudera AI Workbenches.
  3. Click Provision Workbench. The Provision Workbench panel displays.
  4. In Provision Workbench, fill out the following fields.
    1. Workbench Name - Give the Cloudera AI Workbench a name. For example, test-cml. Do not use capital letters in the workbench name.
    2. Select Environment - From the dropdown, select the environment where the Cloudera AI Workbench must be provisioned. If you do not have any environments available to you in the dropdown, contact your Cloudera administrator to gain access.
    3. Namespace - Enter the namespace to use for the Cloudera AI Workbench.
    4. NFS Server - Select Internal to use an NFS server that is integrated into the Kubernetes cluster. This is the recommended selection at this time.
      The path to the internal NFS server is already set in the environment.
  5. In Production Cloudera AI enable the following features.
    1. Enable Governance - Enables advanced lineage and governance features.
      Governance Principal Name - If Enable Governance is selected, you can use the default value of mlgov, or enter an alternative name. The alternative name must be present in your environment and be given permissions in Ranger to allow the Cloudera AI Governance service deliver events to Atlas.
    2. Enable Model Metrics - Enables exporting metrics for models to a PostgreSQL database.
  6. In Other Settings enable the following features.
    1. Enable TLS - Select this to enable https access to the workbench.

      To enable TLS, follow the guidelines in Deploying an Cloudera AI Workbench with support for TLS.

      Upload the certificates to Management Console > Settings > CA Certificates > Miscellaneous. Alternatively, you can configure the Certification Manager to automatically generate certificates for the Cloudera AI infrastructure and workload subdomains. For more details, see Certification Manager service for increased security.

    2. Enable Monitoring - Administrators (users with the EnvironmentAdmin role) can use a Grafana dashboard to monitor resource usage in the provisioned workbench.
    3. Cloudera AI Static Subdomain - This is a custom name for the workbench endpoint, and it is also used for the URLs of models, applications, and experiments. Only one workbench with the specific subdomain endpoint name can be running at a time. You can create a wildcard certificate for this endpoint in advance. The workbench name has this format: <static subdomain name>.<application domain>
  7. Click Provision Workbench. The new workbench provisioning process takes several minutes.

After the workbench is provisioned, you can log in by clicking the workbench name on the Cloudera AI Workbenches page. The first user to log in must be the administrator.

Test backing up of the Cloudera AI Workbench. Ensure that the backup completes successfully, and then ensure you have a process to back up the workbench at regular intervals.