Standing up a foreign catalog through a connection
A foreign catalog in Unity Catalog lets Azure Databricks query and manage data in an external database system—such as Azure SQL Database, PostgreSQL, MySQL, or Snowflake—through Lakehouse Federation, without moving or copying the data. To set it up, you first create a Connection object that stores the external system's authentication and network details, then create a foreign catalog that maps to a database in that external system using the connection.
1 · Learn the must-know
- Lakehouse Federation requires a Unity Catalog-enabled workspace and a Databricks Runtime version that supports the specific connector for the external data source.
- A Connection is a securable object created once (via CREATE CONNECTION or the catalog UI) that stores the host, port, and credentials for the external system; it can be reused by multiple foreign catalogs.
- Creating a Connection requires the CREATE CONNECTION privilege on the metastore, and creating a foreign catalog from a connection requires the CREATE CATALOG privilege plus USE CONNECTION (or CREATE FOREIGN CATALOG) on the connection.
- A foreign catalog mirrors a single database/catalog in the external system, exposing its schemas and tables as read-only objects (write support varies by connector and is more limited).
- Query performance and pushdown behavior depend on the specific connector, so not all operations (e.g., certain aggregations or joins) are pushed down to the source system.
- Network connectivity (e.g., firewall rules, VNet/private endpoints, or IP allowlisting on the external system) must be configured separately so the Databricks classic/serverless compute can reach the external data source.
2 · Check your understanding
A data engineer wants Azure Databricks users to query tables in an on-premises SQL Server instance without copying the data into Delta tables. The engineer plans to use Lakehouse Federation. What is the correct sequence of Unity Catalog objects to create?
What you have tried across DP-750's objectives, not a readiness score.
Set up and configure an Azure Databricks environment15-20% of the exam0 of 13 tried
Secure and govern Unity Catalog objects15-20% of the exam0 of 12 tried
Prepare and process data30-35% of the exam0 of 28 tried
Deploy and maintain data pipelines and workloads30-35% of the exam0 of 24 tried
3 · Keep going
Ready for more? Take a weighted mock or try free practice questions.