Skip to main content

Connect Databricks to your data

Unity Catalog stores the credentials for your external data, so you don't put credentials in notebooks or jobs, and you grant, revoke, and audit access in one place. You connect cloud object storage through a storage credential and an external location, and you connect databases and software as a service (SaaS) applications through a Lakeflow Connect ingestion pipeline.

After you connect your data, build your first pipeline, query the results, and schedule the work with jobs. See Build your first ETL pipeline, Query and explore your data, and Orchestrate workloads with Lakeflow Jobs.

Choose how to connect​

The following table summarizes each approach:

Data source

How you connect

Instructions

Cloud object storage

A Unity Catalog storage credential and external location for reading and writing data in place

Govern access to cloud object storage with Unity Catalog

Databases and SaaS applications

A Lakeflow Connect connection and ingestion pipeline for incremental ingestion into Unity Catalog tables; either a gateway and ingestion pipeline or a single integrated pipeline for change data capture database connectors

Ingest data from databases and SaaS applications

Data source

How you connect

Instructions

Cloud object storage

A Unity Catalog storage credential and external location for reading and writing data in place

Govern access to cloud object storage with Unity Catalog

Databases and SaaS applications

A Lakeflow Connect connection and ingestion pipeline for incremental ingestion into Unity Catalog tables; either a gateway and ingestion pipeline or a single integrated pipeline for change data capture database connectors

Ingest data from databases and SaaS applications

For every way that Databricks connects to external data and services, see Connect to data sources and external services.

Next steps​