To access a Databricks cluster for running jobs, each user of the Trifacta® platform must insert their personal access token into their profile. This configuration enables the user to authenticate to a connected Databricks cluster.
NOTE: Each user must insert a personal access token into their profile. Users that do not provide a personal authentication token cannot run jobs on Databricks, including transformation, sampling, and profiling jobs.
NOTE: When you reset your personal access token (PAT), a new cluster is created if your new token does not have access to your current cluster. If you are resetting an expired personal access token, no new cluster is created. This new cluster is created when you first request access to the Databricks cluster. When you next use an interface that require access to the cluster, such as the relational browser, it may take some time to load.
Acquire your Databricks personal access token.
NOTE: Your Databricks personal access token must be acquired from the same region as your Databricks deployment. This region name is available through the Trifacta application.
For more information, see https://docs.azuredatabricks.net/api/latest/authentication.html#requirements.
Login to the application. From the menu bar, select User menu > Preferences > Databricks.
For Azure developments: Acquire the Azure Databricks personal access token from the same region as your Azure Databricks deployment. The region name is available through the Trifacta application. For more information, see Configure for Azure Databricks.
For AWS developments: Edit the workspace URL, as required and click Save.
- The existing property
databricks.serviceUrlis used to configure the URL to the Databricks Service to run Spark jobs. For more information, see Configure for AWS Databricks.
databricks.serviceUrldefines the default Databricks workspace for all user in the Trifacta Self-Managed Enterprise Edition workspace.
- You can override the default settings in this page.
- The existing property
- Personal access token: In the Personal Access Token field, paste your token.
To use a different token, click Change.
Databricks table cluster name: Each user can specify the name of a cluster to use to browse a Databricks Tables deployment.
NOTE: This cluster must be created and maintained by your Databricks administrator. This cluster can be shared among multiple users.
Databricks policy name: To select the cluster policy to use when you are executing jobs on the cluster, click Edit. The available policies are listed in the drop-down.
NOTE: Cluster policies determine characteristics of Databricks clusters that are used or created for job execution. This feature requires additional configuration.
- Click Save.
This page has no comments.