An analytics team needs a warehouse holding several petabytes of history that many concurrent BI users query. Compute must be pausable and resizable without moving the data. Which service should the team use?
Show answer and explanation
Answer: C
A dedicated SQL pool in Azure Synapse Analytics runs a massively parallel processing engine across distributed compute nodes and keeps data in separate storage. Compute can be scaled or paused without moving the data, and the service is built for very large warehouse tables. Hyperscale targets OLTP-style databases, the Cosmos DB transactional store is not a warehouse engine, and Managed Instance is a lift-and-shift platform for general SQL Server workloads.
Why the other options are wrong
- A. Hyperscale separates compute and storage for databases that grow large, but it targets OLTP-style workloads rather than massively parallel warehouse scans across petabytes.
- B. The transactional store in Cosmos DB serves operational reads and writes. It is not a columnar MPP engine for scanning very large historical tables under heavy BI concurrency.
- D. Business Critical Managed Instance is a SQL Server-compatible platform for migrated instances. It is not a petabyte-scale MPP warehouse with independently scaled and pausable compute.
Exam tip. Massively parallel warehousing with pause and resize points to dedicated SQL pools.