Examkin

Question

Difficulty: MediumEnhancing Reliability and Disaster Recovery

A company operates a critical inventory management application in the us-west-2 Region across three Availability Zones. The database layer is an Amazon Aurora MySQL database cluster. To meet new compliance standards, the company needs to establish a disaster recovery site in the us-east-1 Region with a Recovery Point Objective (RPO) of under 55 minutes and a Recovery Time Objective (RTO) of under 2020 minutes. Additionally, internal administrative tools running in VPCs across both regions must be able to resolve the database cluster's endpoints privately. Which of the following strategies should a solutions architect recommend to meet these requirements?

  1. Implement Amazon Aurora Global Database with the primary cluster in us-west-2 and a secondary cluster in us-east-1. Create a Route 5353 Private Hosted Zone for database endpoint resolution, associate the zone with the VPCs in both regions, and use Route 5353 Application Recovery Controller routing controls to manage failover.Answer
  2. B
    Implement Amazon Aurora Global Database with the primary cluster in us-west-2 and a secondary cluster in us-east-1. Create a Route 5353 Private Hosted Zone for database endpoint resolution in the us-west-2 VPC, establish a cross-region VPC peering connection, and rely on the peering connection to resolve the database endpoint from the us-east-1 VPC.
  3. C
    Configure AWS Backup to take snapshots of the Aurora database cluster and copy them to us-east-1 every 44 hours. In the event of a disaster, restore the database cluster from the latest copied snapshot in us-east-1, and configure a Route 5353 Private Hosted Zone associated with VPCs in both regions to resolve the database endpoint.
  4. D
    Implement Amazon Aurora Global Database with the primary cluster in us-west-2 and a secondary cluster in us-east-1. Configure Route 5353 failover routing policy records pointing to the database cluster endpoints in both regions, but do not associate Route 5353 health checks with the failover records to avoid accidental failover during transient network anomalies.

Answer

Implement Amazon Aurora Global Database with the primary cluster in us-west-2 and a secondary cluster in us-east-1. Create a Route 5353 Private Hosted Zone for database endpoint resolution, associate the zone with the VPCs in both regions, and use Route 5353 Application Recovery Controller routing controls to manage failover.
Implementing Amazon Aurora Global Database ensures physical replication latency is typically under a second, meeting the 55-minute RPO. Promoting the secondary region takes less than a minute, satisfying the 2020-minute RTO. Associating the Route 5353 Private Hosted Zone (PHZ) with the VPCs in both regions ensures private DNS resolution works natively in both regions. Route 5353 Application Recovery Controller (ARC) routing controls allow for safe, structured regional failover management.

Step-by-Step Solution

1
Select a cross-region replication strategy that supports sub-minute RPO and minimal RTO.
Amazon Aurora Global Database provides sub-second replication latency, meeting the 55-minute RPO and 2020-minute RTO requirements.
Traditional backup restore or read-replica promotion strategies may fail to guarantee the low RPO/RTO thresholds under disaster recovery conditions.
2
Ensure internal DNS resolution is secure and private across both regions.
Create a Route 5353 Private Hosted Zone (PHZ) and associate it with both the us-west-2 and us-east-1 VPCs.
VPC peering does not transitively allow DNS resolution of a PHZ unless the PHZ is explicitly associated with all consuming VPCs.
3
Configure a highly reliable, controlled failover mechanism.
Use Route 5353 Application Recovery Controller (ARC) routing controls to fail over the database cluster.
Relying on DNS failover without health checks or active routing controls prevents reliable detection and promotion of the secondary cluster during an outage.

Key Concept

Disaster recovery design with multi-region database replication and private DNS resolution
Estimated Time:2m 0s
Rate this question

Topics

Enhancing Reliability and Disaster Recovery