Create a Databricks connection

Establish a zero copy connection to the Databricks data analytics platform in Zero Copy Connector Hub.

Before you begin

You can optimize queries to Databricks by enabling table statistics. Consult your data source admin to confirm whether table statistics are enabled in Databricks before enabling this option in Zero Copy Connector Hub.

Role required: df_connection_admin

About this task

Work with your data source admin to create a connection to Databricks. For additional information about connecting to Databricks, refer to the Databricks documentation.

Procedure

Navigate to the available primary connectors in Zero Copy Connector Hub in one of the following ways:
- Navigate to All > Zero Copy Connector Hub > Available connectors > Primary connectors.
- Navigate to Admin > Zero Copy Connector Hub > Available connectors > Primary connectors.
- Find the Databricks connector and select Connect.
On the form, fill in the fields.

Configure the authentication method that you want to use with Databricks.

Field	Description
Name and description
Connection label	Unique name for this connection. This helps in identifying the connection within your system.
Connection name	System-generated name based on the Connection label. This field cannot be modified once the connection is established.
Short description	Description of the connection explaining what it is about.
Connection attributes
Catalog name	Catalog name used for organizing and locating data assets in Databricks.
Connection URL	JDBC URL to establish the connection. For example: `jdbc:databricks://:`
Warehouse or compute path	Databricks warehouse or compute path. For example:`/sql/1.0/warehouses/`
Enable table statistics	Option to enable table statistics. Optimize SQL queries using table statistics by selecting this option. Enabling table statistics allows the system to use estimates of stored data such as row count, distinct values, and data size for enhanced query processing. Note: You must ensure that gathering table statistics is enabled in the data source before selecting this option.

Option

Description

Databricks Service Principal

Option to use Databricks as the OAuth provider. Select this option if you're using a Databricks service principal for authentication. 1. Enter the client ID issued by Databricks for authentication. 2. Enter the secret key associated with the client ID.

Azure Service Principal

Option to use Azure Entra ID as the OAuth provider. Select this option if you're using an Azure managed service principal for authentication. This option requires an established a trust relationship between Databricks and Azure Entra ID. 1. Enter the client ID issued by Azure for authentication. 2. Enter the Azure tenant ID where your Databricks application is registered. 3. Enter the secret key associated with the client ID.

Access Token

Option to use a pre-configured OAuth entity profile.

Select this option if you created a record in the Application Registries [oauth_entity] table with a Databricks or IdP service principal for authentication.

This option keeps credentials within the instance and uses the ServiceNow AI Platform OAuth framework for token lifecycle management. For details on creating an OAuth entity profile, see . When configuring the profile, select Client Credentials as the grant type. If your OAuth provider requires scopes, add them on the OAuth Entity Scopes tab. Consult your data source or identity provider documentation for the required scope values.

Select the OAuth entity profile for your Databricks or IdP service principal.

Select Connect.

Result

A test connection is made to the external data source, verifying that the connection details are correct and the data source is accessible.

What to do next

If the connection succeeds, configure data steward access on the Access Control tab. See Manage access to an established connection using roles.

If the connection fails, verify the connection details with your data source administrator and try again.