Databricks Certified Data Engineer - Associate

Explanation

Option E (CREATE TABLE my_table (id STRING, value STRING);) is the correct choice because:

1. Managed Table: When you create a table using just CREATE TABLE without specifying a location in Databricks, it automatically creates a managed table where Spark manages both the data and metadata.

2. DBFS Storage: By default in Databricks, managed tables are stored in DBFS (Databricks File System) under the managed table location.

3. No External Path Needed: Unlike external tables that require explicit path specification, managed tables don't need a PATH option since Spark handles the storage location automatically.

Why other options are incorrect:

• Option A: Uses USING org.apache.spark.sql.parquet with a custom path, which creates an external table where Spark only manages metadata, not the data.
• Option B: CREATE MANAGED TABLE is not valid Spark SQL syntax - the correct syntax is simply CREATE TABLE for managed tables.
• Option C: Same syntax error as Option B - CREATE MANAGED TABLE is not valid.
• Option D: USING DBFS is not valid syntax - DBFS is the underlying storage system, not a table format.

In Databricks, the default behavior for CREATE TABLE without location specification creates a managed table stored in DBFS, meeting all the requirements: Spark manages both data and metadata, and storage is in DBFS._