You operate the same shopping-cart API in East US and West Europe. Each region must write to and read from a local Redis cache with very low latency. If the regions become temporarily isolated, writes should still be accepted in both regions and data should automatically converge when connectivity is restored. You also need automatic failover without promoting a read replica. What should you implement?

The correct solution is to deploy Azure Cache for Redis Enterprise with Active Geo-Replication. This feature provides active-active replication across multiple regions, allowing both regions to accept writes independently even during network partitions. It uses Conflict-free Replicated Data Types (CRDTs) to automatically resolve conflicts and converge data when connectivity is restored, ensuring low latency reads and writes locally. Automatic failover is inherent in an active-active setup. Incorrect options: Redis Premium with Geo-Replication creates a primary/secondary setup, meaning writes would only be accepted by the primary region, failing the requirement for writes in both regions during isolation. Exporting/importing data to Azure Storage is a manual and periodic process, which does not provide real-time synchronization, low latency, or automatic conflict resolution. Using a single Enterprise cache in East US and relying on client retries from West Europe would result in high write latency for the West Europe region and would not allow writes during isolation.