Creating volumes that match how files should be stored
Unity Catalog volumes let you govern and access non-tabular files (e.g., images, CSVs, JSON, models) inside a catalog/schema using the /Volumes/catalog/schema/volume_name path, independent of table storage. You choose between managed volumes (Databricks-managed storage location) and external volumes (existing cloud storage governed via a storage credential and external location) based on where the data should physically reside.
1 · Learn the must-know
- Creating a volume requires CREATE VOLUME privilege on the parent schema plus USE CATALOG and USE SCHEMA privileges, and the workspace must be Unity Catalog-enabled.
- Managed volumes store data under the schema's (or catalog's) default managed storage location and require no LOCATION clause; dropping a managed volume deletes the underlying files.
- External volumes require an existing external location (backed by a storage credential) and are created with CREATE EXTERNAL VOLUME ... LOCATION 'abfss://...'; dropping the volume does not delete the underlying data.
- Volumes are used for unstructured/semi-structured file access (via Volumes File System paths or dbutils/os APIs), not for storing Delta tables—tabular data should still use managed or external tables.
- Volume names must be unique within a schema, and grants (READ VOLUME, WRITE VOLUME, ALL PRIVILEGES) can be applied at the volume level just like tables for fine-grained governance.
- You can create volumes via SQL (CREATE VOLUME), Catalog Explorer UI, or Databricks SDK/REST API, but the underlying external location/storage credential must already exist before creating an external volume.
2 · Check your understanding
A data engineering team has an existing ADLS Gen2 container holding millions of sensor log files already organized by date. They want Unity Catalog to govern read and write access to these files without copying or moving them, and they want the files to remain untouched in storage even if the volume object is later dropped. Which approach satisfies these requirements?
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.