Cloudera Docs

Creating a Workload as a Contributor

Create workloads in Cloudera AI by assigning an Operator-level service account to run jobs, applications, or models under a shared identity instead of your individual user account.

The ability to assign these service accounts through the Run as feature depends directly on the project role of the user and the permission level of the target account. While Site Administrators, Project Admins, and Project Owners can assign both Admin and Operator service accounts, Project Contributors are strictly limited to assigning Operator-level service accounts and are blocked by UI and API validation from selecting administrator-level options.

When using team-level assignments, the machine users available in the Run as dropdown list are determined by the team's project role:

  • Adding a team with an Operator grants Contributors access to all machine users within that team.
  • An Administrator team role restricts visibility to administrators only.
  • Teams configured with the Inherit filter machine users dynamically based on the active user's permissions.
  • Teams designated with the Contributors role do not expose any machine users for workload execution.
  1. In the Cloudera console, click the Cloudera AI tile.
    The Cloudera AI home page is displayed.
  2. Click on the name of the workbench.
    The workbench Home page is displayed.
  3. Click Projects in the left navigation pane.
  4. Click New Job (or Application or Model).
  5. Configure your basic runtime settings, script paths, and engine resources.
  6. Scroll to the Security / Execution Settings section and locate the Run as drop-down list.
  7. Select an available Operator-level service account from the list.
  8. Click Create.
    The workload now deploys and runs using the scoped credentials of the chosen service account without requiring an administrator to provision it for you.