Amazon RDS Read Replicas

Correct Answer: Place the Read Replica in the same AZ as the primary instance

Read latency perceived by the application is dominated by network round-trip time between the application servers and the read replica. In this scenario, the application and the primary DB instance are in the same Availability Zone, while the read replica is in a different AZ. Moving the read replica into the same AZ as the primary also co-locates it with the application servers, minimizing cross-AZ network hops and reducing read latency. Enabling Multi-AZ improves availability but does not serve reads from the standby and will not reduce read latency. Increasing storage capacity targets storage throughput/IOPS, which is not the bottleneck when the issue is network distance. Enabling Performance Insights provides monitoring and diagnostics but does not directly reduce latency.

Correct Answer: Create RDS Read Replicas

Creating RDS Read Replicas helps to distribute read workloads among the replicas, thus reducing the impact of read workloads on the main database. Other options like increasing instance size, changing the database engine, enabling Multi-AZ, or adding more storage space may improve some aspects of performance but would not specifically address the increased read workloads issue. AWS CLI replication is not an optimal solution for handling read workloads in Amazon RDS.

Correct Answer: Use the Read Replica's DNS endpoint in your application

When you have a Read Replica, you can direct your application's read traffic to the Read Replica by using its DNS endpoint. Modifying the security group, enabling Multi-AZ deployment, or using the primary instance's DNS endpoint would not redirect read traffic to the Read Replica. Creating a new RDS instance or modifying instance size wouldn't specifically solve the read performance issue.