Isolating tables in separate databases based on data quality tiers allows for easy permissions management through database ACLs and allows physical separation of default storage locations for managed tables.

Because databases on Databricks are merely a logical construct, choices around database organization do not impact security or discoverability in the Lakehouse.

Storing all production tables in a single database provides a unified view of all data assets available throughout the Lakehouse, simplifying discoverability by granting all users view privileges on this database.

Working in the default Databricks database provides the greatest security when working with managed tables, as these will be created in the DBFS root.

Because all tables must live in the same storage containers used for the database they're created in, organizations should be prepared to create between dozens and thousands of databases depending on their data isolation requirements.