3. Data access and ETL
You'll connect Databricks to your organization's data sources in this section.
Prereqs: Account and workspaces
The big picture
Your workspaces, users, and groups are set up. None of that helps until Databricks can actually reach your data. There are two ways in: direct access to cloud object storage, or ingestion from an external database or SaaS app. This section sets up both.
Starter Journey Progress
Operations and CI/CD
Data science
Genie Ontology
Governance
Data access and ETL
Account and workspaces
IntroductionDEVSTGPRD
Completed Current Pending
How it works
Every path runs through Unity Catalog. An S3 bucket or a PostgreSQL database, it does not matter: each connection is a UC object with its own permissions and audit trail. That is the whole point. You govern access in one place instead of chasing credentials scattered across notebooks.
Which path you take depends on where the data sits:
| Category | Use when |
|---|---|
| Cloud object storage | Your data lives in S3, ADLS, or GCS and you need Databricks to read or write it directly |
| Databases and SaaS ingestion | You need to ingest data from external databases or SaaS platforms into UC tables |
Next
- Do next: Cloud object storage
- Learn why: Unity Catalog foundations
- Reference: Connect to data sources (Databricks docs)