Examkin

Question

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

An e-commerce company hosts its critical shopping cart and order processing application on AWS. The application runs on Amazon ECS tasks using the AWS Fargate launch type in the us-east-1 Region, backed by an Amazon Aurora PostgreSQL Multi-AZ DB cluster. The company needs to design a disaster recovery (DR) solution in the us-west-2 Region. The DR solution must support a Recovery Point Objective (RPO) of 5 minutes and a Recovery Time Objective (RTO) of 15 minutes. The company wants to minimize costs during normal operations while ensuring the compute capacity can scale up rapidly during a failover. Which combination of actions should the Solutions Architect take to meet these requirements? (Select TWO.)

  1. Set up an Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2.Answer
  2. Deploy a Warm Standby compute environment by maintaining a scaled-down ECS service in us-west-2. Configure an Application Load Balancer in us-west-2, and set up Amazon Route 53 failover routing with health checks pointing to both regions.Answer
  3. C
    Create a cross-region read replica of the Aurora PostgreSQL database in us-west-2. In the event of a disaster, configure Amazon Route 53 to automatically promote the read replica to a standalone DB cluster and redirect database traffic to the promoted instance.
  4. D
    Set up a Pilot Light compute environment by maintaining an Amazon ECS task definition and an Application Load Balancer in us-west-2, with the ECS service task count set to 0. Create an AWS CloudTrail log filter that triggers an AWS Lambda function to provision the tasks when us-east-1 is unresponsive.
  5. E
    Establish active-active routing by configuring Amazon Route 53 latency routing to distribute user traffic between us-east-1 and us-west-2. Configure an AWS Lambda function to dynamically adjust weights to 0 for us-east-1 if it goes offline.

Answer

To meet the RPO of 5 minutes and RTO of 15 minutes while minimizing cost, the Solutions Architect should set up an Amazon Aurora Global Database with a secondary cluster in us-west-2, and deploy a Warm Standby compute environment by maintaining a scaled-down ECS service with Route 53 failover routing and health checks in us-west-2.
The correct architecture combines Amazon Aurora Global Database for the data layer and a Warm Standby compute environment for the application layer. Aurora Global Database provides sub-second replication to the secondary region, satisfying the 5-minute RPO, and supports promotion of the secondary cluster to primary in less than a minute, supporting the 15-minute RTO. For the compute layer, keeping a scaled-down ECS service running in the secondary region ensures that network paths, load balancers, and tasks are warm and ready to scale up rapidly using ECS Auto Scaling. Configuring Route 53 failover routing with health checks ensures that user traffic is automatically and dynamically redirected to the secondary region if the primary region's Application Load Balancer becomes unhealthy.

Step-by-Step Solution

1
Evaluate the database replication strategy to satisfy the 5-minute Recovery Point Objective (RPO).
Identify that Amazon Aurora Global Database uses storage-level, physical replication to copy data asynchronously to a secondary region with latency of less than one second, far exceeding the RPO requirement.
Choosing the correct database replication method ensures minimal data loss and compliant RPO.
2
Evaluate the database failover strategy to satisfy the 15-minute Recovery Time Objective (RTO).
Confirm that an Aurora Global Database secondary cluster can be promoted to a standalone read-write cluster in under a minute, meeting the RTO requirement.
Rapid database promotion is critical for minimizing overall application downtime during a regional outage.
3
Determine the optimal compute deployment model to balance the 15-minute RTO and cost-minimization goals.
Select a Warm Standby compute model where a scaled-down ECS service is continuously running in the secondary region. This avoids the long startup latencies of a Pilot Light setup (which starts with 0 tasks) while costing less than an Active-Active deployment.
Maintaining active but scaled-down compute capacity allows tasks to scale up rapidly using ECS Auto Scaling and accept traffic quickly.
4
Establish the DNS routing and health check policy to automate regional failover.
Configure Amazon Route 53 failover routing with health checks targeting both Application Load Balancers. If the us-east-1 load balancer fails, Route 53 automatically redirects client traffic to the us-west-2 load balancer.
Automating traffic redirection based on endpoint health ensures seamless switchover and minimal manual intervention.

Key Concept

Disaster recovery planning using a combination of Aurora Global Database and Warm Standby compute strategies to achieve low RTO and RPO targets cost-effectively.
Rate this question

Topics

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

More questions from this topic