A migrated application stores structured product data and transient user session data. The company wants product data and session data decoupled and wants replication to another AWS Region for disaster recovery. Which design gives the HIGHEST performance?

This option provides the highest performance by separating the data types into services optimized for their specific access patterns and requirements. Amazon RDS with a cross-Region read replica is suitable for structured product data, offering relational database features and disaster recovery. Storing transient user session data in an Amazon DynamoDB global table is ideal because DynamoDB provides high-performance, low-latency access, and automatic replication across regions, which is crucial for session data that needs to be quickly accessible and resilient. Incorrect options: A single Amazon RDS instance with separate schemas for both data types would lead to contention and reduced performance, especially for high-volume session data, and doesn't fully decouple the workloads. Amazon ElastiCache for Memcached is a good caching solution but is not a global datastore for persistent session data that requires cross-Region replication for disaster recovery. It's primarily an in-memory cache, not a durable data store for this use case. While DynamoDB global tables for both product and session data offer high performance and global replication, product data is structured and often benefits from the relational capabilities and strong consistency models of RDS. Using DynamoDB for product data might require more complex application logic to manage relationships compared to RDS.