Databricks Certified Data Engineer - Professional
Get started today
Ultimate access to all questions.
Which approach demonstrates optimal implementation practices for this Lakehouse migration, where bronze tables serve production data engineering, silver tables support both data engineering and machine learning, and gold tables power BI/reports—with PII present across all tiers but properly pseudonymized/anonymized in silver and gold—while balancing security requirements with cross-team collaboration needs?
Which approach demonstrates optimal implementation practices for this Lakehouse migration, where bronze tables serve production data engineering, silver tables support both data engineering and machine learning, and gold tables power BI/reports—with PII present across all tiers but properly pseudonymized/anonymized in silver and gold—while balancing security requirements with cross-team collaboration needs?
Explanation:
Option A is correct because isolating tables into separate databases based on data tiers (bronze, silver, gold) aligns with Databricks best practices. This approach enables efficient permissions management via database ACLs, allowing granular control over access (e.g., restricting bronze tables to data engineers while granting ML teams access to silver and BI teams to gold). Physical separation of storage for managed tables also improves governance and security. Options B, C, D, and E are incorrect: B ignores the security impact of database organization; C risks exposing PII by centralizing access; D misrepresents the default database as secure; and E exaggerates the need for excessive databases. A balances security and collaboration by structuring databases around data tiers.