Databricks Certified Data Engineer - Associate

Explanation

The correct answer is A. Table because:

1. Persistent across sessions: Tables are saved to physical storage (like DBFS, cloud storage) and persist beyond the current Spark session, allowing other data engineers to access them in different sessions.

2. Physical storage requirement: The question explicitly states the data entity "must be saved to a physical location." Tables are physically stored data structures, while views are logical abstractions that don't store data physically.

3. Multi-table composition: The data engineer wants to create a data entity "from a couple of tables," which suggests they need to combine data from multiple source tables into a new persistent structure.

Why other options are incorrect:

• B. Function: Functions are reusable code/logic, not data entities that store or combine data from tables.
• C. View: Views are logical representations that don't store data physically. They are saved as metadata but the underlying data isn't duplicated to a physical location.
• D. Temporary view: Temporary views are session-scoped and don't persist beyond the current session, so they cannot be used by other data engineers in other sessions.

Best Practice: When you need to create a reusable data entity from multiple tables that persists across sessions and is physically stored, you should create a table (either managed or external). This could be done using CREATE TABLE AS SELECT or by writing the results of a query to a table location.