Soru

Zorluk: ZorMulti-AZ, Multi-Region Architectures and Disaster Recovery (DR)

An IoT asset tracking platform for a cold chain logistics company runs its application tier on Amazon EC2 instances in an Auto Scaling group behind an Application Load Balancer in the us-east-1 Region. The database tier uses a Single-AZ Amazon RDS for PostgreSQL DB instance. The company needs to design a disaster recovery (DR) strategy in the us-west-2 Region that achieves a Recovery Point Objective (RPO) of 55 minutes and a Recovery Time Objective (RTO) of 1515 minutes, while minimizing ongoing run costs. Which combination of actions should a solutions architect recommend to meet these requirements? (Select TWO.)

  1. Configure a cross-region read replica for the Amazon RDS for PostgreSQL DB instance in us-west-2 to replicate data asynchronously.Cevap
  2. B
    Configure Amazon RDS Multi-AZ replication directly between us-east-1 and us-west-2 to enable automatic, synchronous database failover to the secondary region.
  3. Pre-deploy the Application Load Balancer in us-west-2, configure an Auto Scaling group with a minimum capacity of 11 instance, and use Amazon Route 53 failover routing with active health checks.Cevap
  4. D
    Deploy an Amazon Route 53 latency routing policy to distribute traffic across both regions, and use AWS CloudFormation to provision the Application Load Balancer and Auto Scaling group in us-west-2 from scratch upon detecting a failure.
  5. E
    Back up the database daily using AWS Backup, copy the recovery points to an Amazon S3 Glacier Flexible Retrieval vault in us-west-2, and restore the database to a new instance during a failover.

Cevap

The correct strategy combines configuring an Amazon RDS for PostgreSQL cross-region read replica in the standby region with pre-deploying the Application Load Balancer and a scaled-down Auto Scaling group (minimum capacity of 11 instance) in that region, combined with Route 53 failover routing and health checks.
To meet both the 55-minute RPO and 1515-minute RTO cost-effectively, a Warm Standby disaster recovery pattern is required. The database tier uses asynchronous cross-region read replication to keep replica lag under a few minutes, meeting the RPO. Promoting the replica to primary takes under 1010 minutes. The compute tier pre-deploys the Application Load Balancer and keeps a minimum of 11 EC2 instance active in the Auto Scaling group, enabling immediate request handling upon failover and rapid scale-out, satisfying the RTO. Route 53 failover routing automatically detects primary region failure and routes traffic to the standby load balancer.

Adım Adım Çözüm

1
Evaluate database disaster recovery options that satisfy the recovery objectives.
Confirm that an Amazon RDS cross-region read replica meets the 55-minute RPO through asynchronous replication, and its promotion time meets the 1515-minute RTO.
Asynchronous replication across regions minimizes performance overhead on the primary database while keeping the standby database updated within seconds or minutes.
2
Determine the minimum compute deployment required in the secondary region to guarantee a failover time of less than 1515 minutes.
Choose a Warm Standby pattern by pre-deploying the Application Load Balancer and running a minimum of 11 EC2 instance in the Auto Scaling group in the recovery region.
Deploying resources from scratch on failover (Pilot Light) takes too long to meet the 1515-minute RTO, whereas keeping a scaled-down compute tier active allows rapid scale-out and immediate request routing.
3
Configure DNS routing to handle failover automated detection.
Implement an Amazon Route 53 failover routing policy using health checks pointed at the primary region's resources.
Active health checks allow Route 53 to automatically detect a regional outage and redirect users to the pre-deployed standby load balancer in the secondary region.

Anahtar Kavram

Warm Standby Disaster Recovery Pattern
Bu soruyu puanla