Examkin

Soru

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

A flight booking aggregator application hosts its primary infrastructure in the us-east-1 Region. The architecture consists of Amazon EC2 instances in an Auto Scaling group behind an Application Load Balancer, and a Multi-AZ Amazon RDS for MySQL database. The company needs to design a cross-region disaster recovery (DR) plan to a secondary AWS region. The DR solution must achieve a Recovery Time Objective (RTO) of 44 hours and a Recovery Point Objective (RPO) of 1515 minutes while minimizing ongoing infrastructure running costs. Which solution meets these requirements?

  1. A
    Configure the RDS database with Multi-AZ deployment and use an Auto Scaling group across multiple Availability Zones in the primary region. Configure Amazon Route 53 latency routing to automatically direct traffic.
  2. Configure a cross-region read replica of the RDS database in the secondary region. Store pre-baked AMIs of the EC2 instances in both regions. In the event of a disaster, promote the read replica to a standalone database and launch the EC2 instances in the secondary region using the AMIs.Cevap
  3. C
    Export database backups every 1515 minutes to Amazon S3 Glacier Flexible Retrieval and configure S3 Cross-Region Replication to copy the backups to the secondary region. In the event of a disaster, retrieve the backups from Glacier to restore the database and launch the EC2 instances.
  4. D
    Deploy a scaled-down active version of the application stack, including a running, smaller RDS database instance and an active EC2 instance in the secondary region, and scale up the environment during a failover event.

Cevap

The correct solution configures a cross-region read replica of the RDS database to meet the RPO and stores pre-baked AMIs to launch EC2 instances in the secondary region during a disaster to meet the RTO while minimizing ongoing costs.
The correct solution uses a Pilot Light disaster recovery strategy. Creating a cross-region read replica of the Amazon RDS database ensures that data is continuously replicated asynchronously to the secondary region, which achieves a Recovery Point Objective (RPO) of minutes, well within the 1515-minute limit. Storing pre-baked AMIs of the EC2 instances allows them to be launched quickly during a failover event, which easily meets the Recovery Time Objective (RTO) of 44 hours. Because the EC2 instances are not running in the secondary region until a disaster occurs, this approach minimizes ongoing infrastructure running costs.

Adım Adım Çözüm

1
Analyze the disaster recovery constraints: RTO must be under 44 hours, RPO must be under 1515 minutes, and ongoing running costs must be minimized.
The RPO constraint requires a data replication mechanism that lag is less than 1515 minutes. The RTO constraint allows up to 44 hours to bring the system online. Cost minimization suggests keeping compute resources offline (Pilot Light) rather than scaled down and active (Warm Standby).
Identifying the target recovery objectives determines the boundary between acceptable DR strategies (Backup & Restore, Pilot Light, Warm Standby, Active-Active).
2
Evaluate database replication options for Amazon RDS for MySQL.
A cross-region read replica replicates data asynchronously, typically within seconds or minutes, satisfying the 1515-minute RPO. Restoring from S3 Glacier backups fails the RTO due to retrieval times.
Ensuring the selected replication technology satisfies the RPO requirement.
3
Evaluate the compute environment configuration for the recovery region.
Storing pre-baked AMIs and launching EC2 instances in the recovery region only when a disaster occurs keeps ongoing compute costs at zero, meeting the cost optimization goal while easily completing the failover within 44 hours.
Selecting the most cost-efficient compute strategy that still satisfies the recovery timeline.

Anahtar Kavram

AWS Disaster Recovery Strategies: Pilot Light vs. Warm Standby
Bu soruyu puanla

Konular

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

Bu konudan daha fazla soru