Integrating with Google Cloud Platform

You can configure an integration with the Google Cloud Platform (GCP) to allow Cloud Pak for Data as a Service users to access data sources from GCP. Before proceeding, make sure you have proper permissions.

After configuring an integration, you’ll see it under Service instances. For example, you’ll see a new GCP tab that lists your BigQuery datasets and Storage buckets.

To configure an integration with GCP:

  1. Log on to the Google Cloud Platform at https://console.cloud.google.com.
  2. Go to IAM & Admin > Service Accounts.

  3. Open your project and then click CREATE SERVICE ACCOUNT.

    Create GCP service account

  4. Specify a name and description for the new service account and click CREATE. Specify other options as desired and click DONE.

    Create GCP service account

  5. Click the actions menu next to the service instance and select Create key. For key type, select JSON and then click CREATE. The JSON key file will be downloaded to your machine.

    Create key

    Select JSON key type

    Important! Store the key file in a secure location.

  6. In Cloud Pak for Data as a Service, under Administrator > Cloud integrations, go to the GCP tab, enable integration, and then paste the contents from the JSON key file into the text field. Only certain properties from the JSON will be stored, and the private_key property will be encrypted.

    Paste GCP key

  7. Go back to Google Cloud Platform and edit the service account you created previously. Add the following roles:

    Add roles to service account

Now you’ll see BigQuery datasets and Storage buckets under Service instances, and they’ll also be available as sources for creating new connections:

Add roles to service account

You can now preview data from GCP data sources, and access GCP data in Notebooks, Data Refinery, SPSS Modeler, and other tools in projects and in catalogs. You can see your GCP instances under Service instances > GCP.

Next steps