Copyright (c) 2026 MindMesh Academy. All rights reserved. This content is proprietary and may not be reproduced or distributed without permission.

5.3.2. Data Ingestion Questions

Question 4

A retail company needs to track customer address changes over time for historical sales analysis by region. When a customer moves, the system should preserve the old address record and create a new one.

Which SCD type should you implement?

  • A. Type 0
  • B. Type 1
  • C. Type 2
  • D. Type 3
Answer: C. Type 2

Explanation: Type 2 SCD preserves complete history by adding new rows when values change. This enables historical analysis showing which region a customer was in at the time of each purchase. Type 1 would overwrite history (inappropriate for historical analysis), Type 0 never updates, and Type 3 only keeps current and previous values.

Question 5

You need to replicate data from Azure Cosmos DB and Azure Databricks Unity Catalog into Microsoft Fabric. Which mirroring approach should you use for each?

  • A. Database mirroring for both
  • B. Metadata mirroring for both
  • C. Database mirroring for Cosmos DB, Metadata mirroring for Databricks
  • D. Metadata mirroring for Cosmos DB, Database mirroring for Databricks
Answer: C. Database mirroring for Cosmos DB, Metadata mirroring for Databricks

Explanation: Azure Cosmos DB uses database mirroring, which replicates actual data using CDC. Azure Databricks uses metadata mirroring, which only replicates Unity Catalog information—the data stays in Databricks storage.

Question 6

You need to load 50 Parquet files into a Fabric Data Warehouse table. The files are located in Azure Blob Storage with names following the pattern sales_*.parquet.

What should you use?

  • A. Individual INSERT statements for each file
  • B. COPY INTO statement with wildcard
  • C. Dataflow Gen2 for each file
  • D. Create a shortcut to the files
Answer: B. COPY INTO statement with wildcard

Explanation: The COPY INTO statement supports wildcards for loading multiple files efficiently in a single operation. Individual INSERT statements would be extremely slow and resource-intensive. Dataflows for each file would be inefficient. Shortcuts provide access but don't load data into tables.

Alvin Varughese
Written byAlvin Varughese
Founder•15 professional certifications