A company must design a disaster-recovery (DR) plan for its ecommerce website. The site runs on t3.large EC2 instances in an Auto Scaling group across multiple Availability Zones and uses an Amazon RDS for MySQL DB instance. In a disaster, traffic must fail over to a secondary Region. The target RPO is 30 seconds and the RTO is 10 minutes. Which solution meets these objectives most cost-effectively?

The correct solution meets the RPO of 30 seconds and RTO of 10 minutes cost-effectively. A cross-Region RDS read replica provides a low RPO for the database. AWS Elastic Disaster Recovery (DRS) continuously replicates EC2 instances, ensuring a low RPO for the application servers. Keeping DR EC2 instances at minimal capacity and scaling up on failover is cost-effective compared to running at full capacity. Route 53 failover routing automates the traffic switch. Incorrect options: Backing up EC2 instances every 30 seconds with AWS Backup and restoring them would likely exceed the 10-minute RTO due to restore times and would be more complex to manage for continuous replication. Manually restoring backups would not meet the 10-minute RTO. Running the Auto Scaling group at full capacity in the DR Region (active-active) is more expensive and not necessary to meet the given RTO/RPO, making it less cost-effective. While an Aurora Global Database offers excellent RPO/RTO, the question asks for the most cost-effective solution for the given requirements, and the RDS read replica combined with DRS is sufficient.