AWS Certified Solutions Architect Professional SAP-C02 Practice Question
A company runs its critical payment-processing application in the us-east-1 Region. The application layer consists of Amazon ECS Fargate tasks behind an Application Load Balancer (ALB). The data layer uses an Amazon Aurora MySQL cluster. The business requires the application to remain available even if the entire us-east-1 Region becomes unavailable. The recovery time objective (RTO) must not exceed 5 minutes, the recovery point objective (RPO) must not exceed 1 minute, and failover must occur automatically while adding as little operational overhead as possible. Which architecture meets these requirements?
Deploy an identical ECS Fargate service and ALB in a second Region, convert the Aurora cluster to an Aurora Global Database with a secondary cluster in that Region, place both ALBs behind an AWS Global Accelerator, and use an Application Recovery Controller Region-switch plan to automatically promote the secondary cluster and redirect traffic during a Regional failure.
Create an Amazon RDS cross-Region read replica of the Aurora cluster in a second Region. Use AWS Systems Manager Automation and AWS CloudFormation to promote the replica and redeploy ECS tasks when the primary Region fails, then update a weighted Amazon Route 53 record to send traffic to the new ALB.
Configure AWS Elastic Disaster Recovery (AWS DRS) to continuously replicate the ECS tasks and Aurora database volumes to a staging subnet in another Region. When a disruption is detected, start a recovery job with AWS DRS and manually update DNS to restore service.
Enable Multi-AZ for the Aurora cluster and spread the ECS service across three Availability Zones behind the existing ALB, relying on ALB cross-zone load balancing to maintain availability if an outage occurs.