Examkin

Question

Difficulty: MediumMulti-AZ, Multi-Region Architectures and Disaster Recovery (DR)

A digital music streaming provider hosts its platform backend in the us-east-1 Region. The architecture consists of an Application Load Balancer (ALB) and Amazon EC2 instances in an Auto Scaling group, with data stored in an Amazon RDS for PostgreSQL DB instance. The provider wants to implement a disaster recovery (DR) strategy in the us-west-2 Region to achieve a Recovery Time Objective (RTO) of 15 minutes and a Recovery Point Objective (RPO) of 5 minutes. The solution must minimize ongoing infrastructure costs in the secondary region.

Which configuration will meet these requirements?

  1. Configure a cross-region read replica of the RDS DB instance in us-west-2. Deploy a pilot light environment in us-west-2 with an Auto Scaling group set to a minimum capacity of 0. Write a script to promote the read replica, scale up the Auto Scaling group, and configure Amazon Route 53 failover routing.Answer
  2. B
    Configure a cross-region read replica of the RDS DB instance in us-west-2. Deploy a warm standby environment in us-west-2 with an Auto Scaling group running at a reduced capacity of 2 instances. Configure Amazon Route 53 latency routing to route traffic to the region with the lowest latency.
  3. C
    Create hourly snapshots of the RDS DB instance and copy them to us-west-2. Store the snapshots in Amazon S3 Glacier Flexible Retrieval. Deploy a pilot light environment in us-west-2 with an Auto Scaling group set to a minimum capacity of 0. In a failover event, restore the database from the S3 Glacier snapshot.
  4. D
    Deploy a Multi-AZ DB cluster for the RDS PostgreSQL DB instance with the primary in us-east-1 and a standby replica in us-west-2. Configure an Auto Scaling group in us-west-2 to automatically launch EC2 instances when the primary region fails. Configure Amazon Route 53 failover routing.

Answer

Configure a cross-region read replica of the RDS DB instance in us-west-2, deploy a pilot light environment in us-west-2 with an Auto Scaling group capacity of 0, and use automated scripts to promote the replica, scale up the compute resources, and update Route 53 failover routing.
The correct option uses an Amazon RDS cross-region read replica to satisfy the 5-minute RPO since database replication is asynchronous and near real-time. Deploying a pilot light environment with an Auto Scaling group capacity of 0 ensures that no EC2 instances are running during normal operations, minimizing ongoing costs. In the event of a failover, promoting the replica and scaling up the Auto Scaling group via automated scripts can easily be completed within the 15-minute RTO. Finally, Amazon Route 53 failover routing ensures traffic is redirected to the disaster recovery region.

Step-by-Step Solution

1
Select the database replication method that satisfies the 5-minute RPO.
Identify that cross-region read replicas use continuous asynchronous replication, keeping data lag to a few seconds, which satisfies the RPO. Hourly snapshots fail the RPO requirement.
We must ensure database changes are copied to the secondary region fast enough to meet the 5-minute RPO threshold.
2
Determine the most cost-effective compute deployment in the recovery region.
Choose a Pilot Light configuration where the Auto Scaling group has a minimum capacity of 0. This avoids paying for running EC2 instances during normal operations.
The scenario requires minimizing ongoing infrastructure costs, and a Pilot Light environment is cheaper than a Warm Standby.
3
Establish the failover mechanism to meet the 15-minute RTO.
Utilize Amazon Route 53 failover routing to automatically detect primary region failure. Write automated scripts to promote the cross-region read replica to a standalone database and scale the EC2 Auto Scaling group from 0 to its target capacity.
Manual promotion and scaling would take too long, so automation is necessary to guarantee recovery within 15 minutes.

Key Concept

Cross-Region Disaster Recovery (DR) Strategies (Pilot Light vs. Warm Standby) and RDS Replication
Rate this question

Topics

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

More questions from this topic