How do Lakebase and Unity Catalog work together for governed operational applications?

Summary

• Two layers, one platform. Lakebase is Databricks' managed Postgres database for operational and transactional workloads, and Unity Catalog is the platform's unified governance layer. Together they let you build operational applications whose data is discoverable, access-controlled, and lineage-tracked. See the Lakebase documentation.
• Register Lakebase in Unity Catalog. Registering a Lakebase Postgres database in Unity Catalog creates a read-only catalog that mirrors your Postgres schemas, tables, and views, so the database is discoverable in Catalog Explorer and Unity Catalog permissions govern who can view and query it. See Register a Lakebase database in Unity Catalog.
• Synced tables move governed data into the operational layer. Managed Lakeflow pipelines keep a copy of lakehouse (Delta) data in Lakebase Postgres — in Snapshot, Triggered, or Continuous mode — so applications read fresh, enriched data with transactional guarantees. See Sync data to Lakebase.
• Cross-source queries and lineage. Once registered, you can join transactional Postgres tables with Delta tables in a single query, and Unity Catalog provides lineage across operational and analytical data.
• Native support in Databricks Apps. When you add Lakebase as a resource to a Databricks App, Databricks provisions a service principal, grants it a matching Postgres role, and injects connection details — so your app connects without managing credentials. See Connect a Databricks App to Lakebase.

How do Lakebase and Unity Catalog work together for governed operational applications?

Operational applications need a transactional database for fast reads and writes, and governance so the data stays discoverable, access-controlled, and auditable. On the Databricks Data Intelligence Platform, Lakebase provides a managed Postgres database for operational and transactional workloads, and Unity Catalog provides unified governance. They work together so that the same data your applications transact on can be governed and discovered alongside the rest of your lakehouse, and so lakehouse data can be served back into your operational apps.

Why Databricks Lakebase and Unity Catalog for governed operational applications

• Register Lakebase databases in Unity Catalog. Registering a Lakebase Postgres database in Unity Catalog creates a read-only Unity Catalog catalog that mirrors the Postgres database structure — schemas, tables, and views. The database then appears in Catalog Explorer for discovery, and Unity Catalog permissions govern who can view and query it through Unity Catalog and SQL warehouses. See Register a Lakebase database in Unity Catalog.
• Serve lakehouse data with synced tables. A synced table keeps a managed copy of Unity Catalog (Delta) data — such as gold tables, engineered features, or model outputs — in a read-only Postgres table your application can query. Synced tables use managed Lakeflow pipelines to keep both the Unity Catalog synced table and the Postgres table up to date, in one of three modes: Snapshot (one-off), Triggered (scheduled incremental), or Continuous (within seconds of the source). See Sync data to Lakebase.
• Transactional reads and writes at scale. Lakebase Postgres supports up to 1,000 concurrent connections with transactional guarantees, so applications can read enriched, synced data while handling inserts, updates, and deletes in the same database.
• Cross-source queries and end-to-end lineage. Because a registered Lakebase database is visible in Unity Catalog, you can join your application's transactional tables with Delta tables in a single query — combining operational and analytical data without switching contexts — and Unity Catalog tracks lineage across the pipelines that move data between the lakehouse and Lakebase.
• Native support in Databricks Apps. Databricks Apps support Lakebase as a resource type. When you add Lakebase to an app, Databricks creates a service principal for the app, grants it a matching Postgres role, and injects connection details as environment variables, so the app connects to a fully managed Postgres database without handling credentials or connection strings. See Connect a Databricks App to Lakebase.
• Online feature and model serving. With Lakebase as the online store and the lakehouse as the offline store, applications can serve machine learning features and predictions with low latency while the analytical layer stays the source of truth for training and analysis.
• Operational maturity of the platform. Lakebase inherits platform capabilities including built-in monitoring of key database metrics — such as transactions per second, open connections, and resource utilization — and enterprise network security features including PrivateLink and IP access control lists. See Announcing Lakebase.

Getting started

• Start with the Lakebase documentation for an overview of Lakebase instances and projects.
• Register a Lakebase database in Unity Catalog to make it discoverable and governed.
• Use synced tables to serve lakehouse data into your operational application.
• Build an app with Databricks Apps and Lakebase.

FAQs

What happens when I register a Lakebase database in Unity Catalog?

Registration creates a read-only Unity Catalog catalog that mirrors the Postgres database's schemas, tables, and views. The database becomes discoverable in Catalog Explorer, and Unity Catalog permissions govern who can view and query it through Unity Catalog and SQL warehouses.

What are synced tables in Lakebase?

Synced tables keep a managed copy of Unity Catalog (Delta) data in a read-only Postgres table using managed Lakeflow pipelines. They can run in Snapshot, Triggered, or Continuous mode so your application always reads current lakehouse data.

Can I join operational and analytical data in one query?

Yes. After registering a Lakebase database in Unity Catalog, you can join its transactional Postgres tables with Delta tables in a single query, combining operational and analytical data without switching systems.

How does a Databricks App connect to Lakebase?

When you add Lakebase as a resource, Databricks creates a service principal for the app, grants it a matching Postgres role, and injects connection details as environment variables, so the app connects to a fully managed Postgres database without managing credentials.