Skip to main content

GCP manual workspace creation

You'll create a Databricks workspace on GCP in ~10 min.

Prereqs: GCP project, Databricks account console

What you'll walk away with​

A running Databricks workspace in your GCP project, created from the Databricks account console. The wizard handles the service account, GCS storage, and workspace configuration for you.

Prerequisites​

  • A GCP project with Owner permissions.
  • A Databricks account with account-admin privileges.
  • The GCP project the workspace belongs to (for example, the dev workspace in the dev project).

Steps​

1. Create the development workspace​

Follow the video below. As you deploy, pick the right GCP project for the environment.

warning

Deploy each workspace into the matching GCP project. A dev workspace in a production project causes confusion and makes cleanup harder.

2. Repeat for staging and production​

Run the same process for each additional workspace, picking the matching GCP project each time.

Verify​

  1. Log in to the Databricks account console.
  2. Open Workspaces and confirm the new workspace shows Running.
  3. Click the workspace URL and confirm the landing page loads.

Where people trip​

Workspace stuck in provisioning

Check the GCP console for API enablement issues. The workspace needs several GCP APIs enabled on the project: Compute Engine, Cloud Storage, and IAM.

Permission errors during creation

Confirm the authenticated account has Owner on the target GCP project. Viewer and Editor are not enough for the initial workspace creation.

Next​