AWS Certified Solutions Architect – Professional — Question 407

A company is running an email application across multiple AWS Regions. The company uses Ohio (us-east-2) as the primary Region and Northern Virginia (us- east-1) as the Disaster Recovery (DR) Region. The data is continuously replicated from the primary Region to the DR Region by a single instance on the public subnet in both Regions. The replication messages between the Regions have a significant backlog during certain times of the day. The backlog clears on its own after a short time, but it affects the application's RPO.
Which of the following solutions should help remediate this performance problem? (Choose two.)

Answer options

• A. Increase the size of the instances.
• B. Have the instance in the primary Region write the data to an Amazon SQS queue in the primary Region instead, and have the instance in the DR Region poll from this queue.
• C. Use multiple instances on the primary and DR Regions to send and receive the replication data.
• D. Change the DR Region to Oregon (us-west-2) instead of the current DR Region.
• E. Attach an additional elastic network interface to each of the instances in both Regions and set up load balancing between the network interfaces.

Correct answer: C, E

Explanation

Using multiple instances (Option C) allows the replication workload to be distributed and processed in parallel, preventing a single instance from becoming a bottleneck during peak hours. Attaching an additional elastic network interface (Option E) increases the network throughput capacity for each instance, helping to handle the increased traffic volume. Other options, such as moving the DR region to Oregon (Option D), would increase latency and worsen the replication lag.