CERTIFIED-DATA-ANALYST-ASSOCIATE Question #89: Real Exam Question with Answer & Explanation

Question

Which example of data projects represents a common analytics application to be completed in Databricks SQL?

Options

• ATesting the latency of real-time streaming workloads
• BPerforming analysis-specific ETL on gold-layer tables
• CPredicting customer churn by automatically developing time-series models
• DMonitoring the performance of deployed machine learning solutions

Explanation

Option B is correct because Databricks SQL is specifically designed for running SQL-based analytics queries against structured data, and performing ETL transformations on gold-layer tables (the curated, business-ready layer in the Medallion architecture) is a classic BI/analytics workload that fits naturally in this context.

• A is wrong because latency testing of real-time streaming workloads belongs in Spark Structured Streaming or Delta Live Tables, not Databricks SQL, which is optimized for batch analytical queries.
• C is wrong because auto-generating time-series models for churn prediction is a machine learning automation task, better handled by tools like AutoML in Databricks ML, not SQL.
• D is wrong because monitoring deployed ML models falls under MLflow model monitoring or observability tooling, not Databricks SQL.

Memory tip: Think of Databricks SQL as your "analytics endpoint" - it serves business users and analysts writing SQL against clean, finalized data. If the task involves streaming, ML training, or model ops, you've left SQL territory and entered Spark/ML territory.

No community discussion yet for this question.