Microsoft Certified: Fabric Analytics Engineer AssociateImplement and manage semantic models (30-35%)Medium
A data engineer is designing a semantic model in Microsoft Fabric. The model will integrate data from an Azure Data Lake Storage Gen2 (ADLS Gen2) account containing Parquet files. The business requires near real-time analytics on this data, but also needs to support complex aggregations and DAX calculations that perform best with in-memory caching. Which storage mode should the engineer choose for the tables sourced from ADLS Gen2 to meet both requirements?
- ADual
- BImport
- CDirect Lake
- DDirectQuery
Show answer & explanationAnswer & explanation
Correct answer: A. Dual
Dual mode allows tables to behave as both Import and DirectQuery. For tables sourced from ADLS Gen2, Dual mode would enable Direct Lake behavior for near real-time queries while also caching data in memory for optimal performance of complex DAX calculations and aggregations, providing the best of both worlds.
Why the other options are wrong
- B. Import mode provides in-memory performance but requires scheduled refreshes, not near real-time.
- C. Direct Lake is excellent for near real-time from OneLake, but for complex aggregations, caching (like in Import mode) provides superior performance. Dual combines Direct Lake and Import effectively.
- D. DirectQuery provides real-time data but can be slower for complex aggregations as it queries the source directly.
Dual Storage Mode
A table storage mode in Microsoft Fabric that combines aspects of Import and DirectQuery/Direct Lake, allowing the engine to choose the most efficient mode for each query.
- Tables can be queried in DirectQuery/Direct Lake mode for real-time access.
- Data is also cached in memory (like Import) for complex, high-performance queries.
- Provides flexibility and optimizes performance across diverse query patterns.
Memory trick: Dual mode gives you real-time and cached power.