Databricks Data Engineer Associate Question 527
Select 2A company uses Databricks Unity Catalog to manage data access across its organization. The company has multiple business units, including Sales, Marketing, and Finance. Which of the following catalog design practices best supports the segregation of business units while ensuring effective data governance?
- A
Create separate catalogs for each business unit (e.g., Sales_Catalog, Marketing_Catalog, Finance_Catalog).
- B
Store all data for all business units in a single catalog and manage access through table-level permissions.
- C
Create a single catalog and use schema-level segregation for different business units (e.g., Sales_Schema, Marketing_Schema, Finance_Schema).
- D
Use workspace-level segregation by creating separate Databricks workspaces for each business unit.
- E
Use data tagging and classification only to segregate data between business units.
Show answer and explanation
Correct answers: A, D
Explanation
The best practices for segregating business units across catalogs in Databricks involve creating separate catalogs for each business unit and leveraging workspace-level segregation where necessary. These approaches provide robust isolation, simplify governance, and reduce the risk of access control misconfigurations. Schema-level segregation or table-level permissions within a single catalog may be insufficient for large organizations with strict compliance and governance requirements.
- A. Correct.
Creating separate catalogs for each business unit ensures clear segregation of data and simplifies access control and auditing. This is a best practice for organizations with distinct business unit boundaries.
- B. Incorrect.
Storing all data in a single catalog with table-level permissions can lead to complex access control policies and increases the risk of misconfigurations, making it less effective for segregation.
- C. Incorrect.
Using schema-level segregation within a single catalog may work for smaller organizations but does not provide the same level of isolation as separate catalogs, especially for large enterprises with strict governance requirements.
- D. Correct.
Workspace-level segregation is a common practice in Databricks to ensure isolation between different business units. Each workspace can have its own catalog, ensuring clear boundaries and governance.
- E. Incorrect.
Data tagging and classification are useful for identifying and categorizing data but are not sufficient by themselves for ensuring segregation between business units.