AWS Certified Solutions Architect Professional SAP-C02 Practice Question

The correct approach successfully combines services to meet the stringent RPO and RTO requirements while remaining cost-effective. Using an Amazon RDS for PostgreSQL cross-region read replica provides asynchronous replication to the DR region, which is essential for achieving an RPO of less than 1 minute. During a disaster, this replica can be promoted to a standalone, writable database instance in a few minutes, which helps meet the RTO of less than 15 minutes.

For the application tier, a 'pilot light' strategy is the most cost-effective approach. This involves replicating the Amazon Machine Image (AMI) to the DR region and maintaining an Auto Scaling group with a minimum and desired capacity of 0 or 1. In a disaster, this can be quickly scaled out. For data in Amazon S3, enabling Cross-Region Replication (CRR) ensures that objects are automatically and asynchronously copied to a bucket in the DR region, which is suitable for file-based data. Finally, Amazon Route 53 with health checks and a failover routing policy should be used to redirect user traffic to the DR region's resources once they are promoted and scaled.

Using AWS Backup with continuous backups for RDS typically provides an RPO of around 5 minutes, which does not meet the sub-1-minute requirement. Using AWS Database Migration Service (DMS) for DR adds unnecessary complexity compared to native RDS replication features and suggesting a warm standby for EC2 is less cost-effective than a pilot light approach. AWS Elastic Disaster Recovery (DRS) is designed for block-level replication of servers like EC2 or on-premises machines and is not the appropriate or recommended tool for managed database services like Amazon RDS.