A SysOps administrator needs to design a disaster recovery (DR) plan for an application on AWS. The application runs on Amazon EC2 instances behind an Application Load Balancer (ALB). The instances…

Question

Options

• AConfigure Aurora backups to be exported to the DR Region.
• BConfigure the Aurora cluster to replicate data to the DR Region by using the Aurora global
• CConfigure the DR Region with an ALB and an Auto Scaling group. Use the same configuration as
• DConfigure the DR Region with an ALB and an Auto Scaling group. Set the Auto Scaling group's
• EManually launch a new ALB and a new Auto Scaling group by using AWS CloudFormation during

• A
  18% (8)
• B
  67% (30)
• D
  11% (5)
• E
  4% (2)

Explanation

B and C work together to satisfy both the 15-minute RPO and RTO requirements cost-effectively. Aurora Global Database (B) replicates data to the DR Region with sub-second lag, enabling a managed failover in roughly 1 minute - well within the 15-minute RPO. Pre-provisioning an ALB and Auto Scaling group in the DR Region with a warm standby (minimum instances already running) means compute capacity is available immediately on failover, meeting the 15-minute RTO without overprovisioning.

Why the distractors fail:

• A (S3 backup exports) targets RPO, but restoring a database from an S3 export takes hours - far exceeding 15 minutes.
• D likely sets the Auto Scaling group minimum to 0 (cold standby), meaning instances must launch from scratch during a disaster, which risks exceeding the 15-minute RTO.
• E (manually running CloudFormation during an incident) introduces human latency and operational risk, making it nearly impossible to consistently meet a 15-minute RTO.

Memory tip: Think "Global replication + Warm standby = Fast DR." Aurora Global = near-zero RPO; pre-warmed EC2 capacity = fast RTO. If something requires manual steps or a full restore during the incident, it's too slow for a sub-30-minute objective.

No community discussion yet for this question.