AZ-305 Design data storage solutions Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✓

  Azure Cosmos DB with SQL API and multiple write regions

  Why this is correct

  Azure Cosmos DB with the SQL API and multiple write regions is the only option that enables true multi-region writes, allowing every regional replica to accept write operations simultaneously. This is essential for a global real-time collaboration platform because users in different parts of the world experience low-latency writes without being forced to a single primary. Cosmos DB automatically handles conflict resolution using policies such as last-writer-wins or custom merge procedures, and its turnkey global distribution provides high availability (99.999% SLA) and multiple well-defined consistency levels, making it the ideal underlying data store for such a workload.

• ✗

  Azure SQL Database with active geo-replication

  Why it's wrong here

  Azure SQL Database with active geo-replication only provides a readable secondary in other regions; all write operations must still be sent to the primary region. This means remote users would incur cross-region network latency for every write, undermining real-time collaboration responsiveness. Configured correctly, active geo-replication supports failover for disaster recovery, but it does not support multi-region writes or automatic conflict resolution, so it cannot meet the requirement of a globally distributed write-heavy application.

• ✗

  Azure Table Storage

  Why it's wrong here

  Azure Table Storage is a simple NoSQL key-value store that supports only limited indexing (essentially on partition key and row key) and lacks a rich query engine. It does not support multi-region writes, has no automatic conflict resolution mechanism, and does not offer the low-latency, multi-region consistency guarantees required for real-time collaboration. While it can scale elastically, it is designed for cost-effective storage of structured data, not for globally distributed, conflict-ridden collaborative workloads.

• ✗

  Azure Cache for Redis

  Why it's wrong here

  Azure Cache for Redis is an in-memory data store meant for caching and session management, not a fully managed database with durable storage and global replication for writes. Even with Redis persistence or geo-replication (in Redis Enterprise tiers), it is not designed to serve as the authoritative store for a collaboration platform with conflict resolution and multi-region write semantics. Its primary purpose is to accelerate reads by sitting in front of a durable database, so it cannot alone fulfill the demands of a real-time multi-region collaboration solution.