Examkin

Soru

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

A food delivery platform hosts its order dispatch application in the us-west-2 Region. The application tier runs on Amazon EC2 instances in an Auto Scaling group behind an Application Load Balancer, and the database tier uses an Amazon RDS for PostgreSQL Multi-AZ DB instance. The company wants to establish a disaster recovery (DR) solution in the us-east-1 Region. The solution must achieve a Recovery Time Objective (RTO) of 15 minutes and a Recovery Point Objective (RPO) of 5 minutes, while keeping ongoing standby costs as low as possible. Which disaster recovery strategy should a solutions architect recommend to meet these requirements?

  1. A
    Deploy an Auto Scaling group with a desired capacity of 0 EC2 instances behind an Application Load Balancer in the destination region. Create an Amazon RDS cross-region read replica in the destination region. Configure Route 53 failover routing with health checks. In the event of a disaster, promote the read replica to a primary DB instance and increase the desired capacity of the Auto Scaling group.
  2. Deploy a scaled-down Auto Scaling group of EC2 instances behind an Application Load Balancer in the destination region. Create an Amazon RDS cross-region read replica in the destination region. Configure Route 53 failover routing with health checks. In the event of a disaster, promote the read replica to a primary DB instance and scale up the Auto Scaling group.Cevap
  3. C
    Create an Amazon RDS cross-region read replica in the destination region. Configure Route 53 latency routing to distribute traffic between the two regions. Rely on AWS to automatically promote the cross-region read replica to the primary DB instance if the primary region experiences an outage.
  4. D
    Configure daily automated database snapshots and copy them to Amazon S3 Glacier Flexible Retrieval in the destination region. Use an AWS CloudFormation template to deploy the application infrastructure. In the event of a disaster, restore the database from the snapshot in S3 Glacier and run the CloudFormation template.

Cevap

Deploy a scaled-down Auto Scaling group of EC2 instances in the destination region, create an Amazon RDS cross-region read replica, configure Route 53 failover routing, and promote the replica and scale up the instances during a disaster.
The correct strategy is a Warm Standby deployment. By keeping a scaled-down Auto Scaling group of EC2 instances running in the destination region, the application can start serving traffic immediately upon failover and then scale out. The Amazon RDS cross-region read replica maintains a near real-time copy of the database, ensuring the 5-minute RPO is met. During a disaster, the replica is promoted to primary and the EC2 instances are scaled up, which can be accomplished within the 15-minute RTO.

Adım Adım Çözüm

1
Evaluate RPO requirements against replication options.
An RPO of 5 minutes requires continuous replication. An Amazon RDS cross-region read replica provides asynchronous replication with lag typically under a few seconds, meeting the 5-minute RPO.
Backup-and-restore or snapshot copy methods cannot guarantee a 5-minute RPO due to the frequency of snapshots.
2
Evaluate RTO requirements against compute standby options.
An RTO of 15 minutes is very tight. Bootstrapping new EC2 instances from scratch (desired capacity 0) takes time to launch, run user data scripts, and pass load balancer health checks. Keeping a scaled-down Auto Scaling group with at least one active instance allows immediate traffic handling while the group scales up.
Distinguishing between Warm Standby (scaled-down active instances) and Pilot Light (no running instances) is critical for meeting short recovery times.
3
Determine the routing and failover mechanics.
Configure Route 53 failover routing with health checks to detect primary region failure. In a disaster, the read replica must be promoted manually or via automation, and the standby Auto Scaling group must be scaled up to handle production traffic.
RDS cross-region replication is not automatic and requires promotion to become a read-write primary database.

Anahtar Kavram

Warm Standby Disaster Recovery Strategy
Tahmini Süre:2m 0s
Bu soruyu puanla

Konular

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

Bu konudan daha fazla soru