Examkin

Question

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

A collaborative online whiteboard platform runs its production environment in the us-east-1 Region. The architecture includes an Application Load Balancer, an Auto Scaling group of Amazon EC2 instances, and an Amazon RDS for PostgreSQL database. The company needs to design a disaster recovery (DR) plan in the us-west-2 Region. The design must meet a Recovery Time Objective (RTO) of 4 hours and a Recovery Point Objective (RPO) of 15 minutes while minimizing ongoing infrastructure costs.

Which strategy should a solutions architect recommend to meet these requirements?

  1. A
    Establish a backup and restore recovery strategy by taking daily database snapshots, copying them to an Amazon S3 bucket in the secondary Region with an S3 Lifecycle policy to transition them to Amazon S3 Glacier Flexible Retrieval, and deploying compute resources using CloudFormation.
  2. Establish a pilot light recovery strategy by creating an Amazon RDS cross-region read replica in the secondary Region, and using AWS CloudFormation templates to deploy the Application Load Balancer and Auto Scaling group only when a failover is initiated.Answer
  3. C
    Establish a warm standby recovery strategy by running a scaled-down Auto Scaling group and Application Load Balancer in the secondary Region, and maintaining an Amazon RDS cross-region read replica to handle database replication.
  4. D
    Establish an active-passive failover strategy by configuring Amazon Route 53 DNS failover to route traffic to an Amazon RDS cross-region read replica in the secondary Region, which will automatically promote itself and launch the required EC2 instances when the primary Region becomes unhealthy.

Answer

Establish a pilot light recovery strategy by creating an Amazon RDS cross-region read replica in the secondary Region, and using AWS CloudFormation templates to deploy the Application Load Balancer and Auto Scaling group only when a failover is initiated.
The pilot light strategy is the most cost-effective option that meets the RTO of 4 hours and RPO of 15 minutes. An Amazon RDS cross-region read replica provides asynchronous replication with lag typically measured in seconds, satisfying the 15-minute RPO. Since the RTO is 4 hours, there is sufficient time to use AWS CloudFormation to provision the Application Load Balancer and the Auto Scaling group of EC2 instances when a disaster is declared, avoiding the ongoing compute costs of a warm standby.

Step-by-Step Solution

1
Analyze the RTO and RPO requirements.
The RTO is 4 hours, which allows sufficient time to deploy compute resources on demand. The RPO is 15 minutes, which requires continuous database replication rather than daily backup copies.
This establishes the technical constraints for computing (dynamic deployment is allowed) and data (replication must be near real-time).
2
Evaluate the cost constraint.
To minimize ongoing infrastructure costs, the standby region should run as few active compute resources as possible during normal operations.
Running compute instances (like in a warm standby or active-active setup) increases ongoing costs, which must be avoided.
3
Select the optimal disaster recovery pattern.
A pilot light pattern keeps the database replicated live using a cross-region read replica, but leaves the application and load balancing tiers unprovisioned until failover, using CloudFormation to build them when needed.
This meets both the 15-minute RPO and 4-hour RTO while achieving the lowest possible idle infrastructure costs.

Key Concept

Disaster recovery strategies differ in RTO, RPO, and cost; pilot light minimizes costs by only keeping data replication active and provisioning compute resources on demand.
Rate this question

Topics

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

More questions from this topic