Configure zero-copy integration from Databricks using APIs

Learn how to configure a Databricks connection and, when required, interaction mappings for zero-copy integration.

Configure the Databricks connection used by Zero-copy integration. Create the connection and store the Databricks service principal credentials, then test the connection. Create interaction mappings only when you use interaction integration or include interaction attributes in segment rules.

The following roles configure or manage zero-copy integration from Databricks using APIs:

Prerequisites

Before you begin, confirm the following connection requirements are met:

Zero copy integration is enabled for your tenant. For segmentation, the zero_copy_segmentation_enabled feature is enabled. Contact Reltio Support if zero copy integration is not enabled
A Reltio role that allows you to manage data integrations, such as ROLE_ADMIN.
A running Databricks SQL Warehouse.
A Databricks service principal with an OAuth client ID and client secret.
The service principal has CAN USE permission on the SQL warehouse.
The Databricks workspace URL and warehouse ID for the source data.

If you want to map interaction data, also confirm the following requirements are met:

The Reltio tenant contains the interaction type that you want to map.
The service principal has USE CATALOG on the source catalog.
The service principal has USE SCHEMA on the source schema.
The service principal has SELECT on the source table or view.

To configure zero-copy integration from Databricks using APIs

Create a Databricks connection and store OAuth credentials: Create the connection configuration and store the service principal client ID and client secret in Reltio. Reltio uses these credentials to request and refresh Databricks access tokens.

Note: The Databricks connection uses OAuth 2.0 machine-to-machine authentication with a service principal. Personal access Tokens are not supported for this connection.
Test a Databricks connection: Verify that Reltio authenticates to the Databricks workspace and runs a query on the configured SQL warehouse.

The connection test verifies authentication and warehouse access. It does not verify access to mapped tables, views, or segmentation data.
Map a Databricks source object when the capability uses interaction data: Create a mapping when you use interaction integration or include interaction attributes in segment rules. The mapping identifies the Databricks table or view, the interaction attributes to return, and the crosswalk columns that link Databricks rows to Reltio profiles.

Skip this task when segmentation uses profile attributes only.

Result

The Databricks connection is configured and tested. When you create an interaction mapping, the mapped Databricks source is available for profile interactions and interaction-based segment rules.

Segmentation requires Reltio Data Pipeline for Databricks and the permissions described in Configure Databricks permissions for zero-copy segmentation before the first segment evaluation.

Unify and manage your data

Configure zero-copy integration from Databricks using APIs

Footer