Question

Difficulty: HardHigh Availability and Disaster Recovery Design

A financial services company is designing a new multi-region disaster recovery (DR) architecture for a mission-critical billing application. The primary workload runs in a VPC in us-east-1, and the secondary disaster recovery site is located in us-west-2. The application utilizes Amazon Aurora PostgreSQL. The business requires a Recovery Time Objective (RTO) of 10 minutes and a Recovery Point Objective (RPO) of 1 minute. Client application servers in both regions must access the database using a single, consistent private DNS name. Which two database and routing configurations should a Solutions Architect implement to meet these requirements?

  1. Deploy an Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2, utilizing physical replication to meet the 1-minute RPO target.Answer
  2. Create a Route 53 Private Hosted Zone (PHZ) containing a CNAME record pointing to the active database endpoint, and associate this PHZ with the VPCs in both us-east-1 and us-west-2.Answer
  3. C
    Configure Route 53 with a Geolocation routing policy using public DNS records to resolve the database endpoint, without associating the Private Hosted Zone with the us-west-2 VPC.
  4. D
    Set up a standard Amazon Aurora PostgreSQL DB cluster in us-east-1 with cross-region read replicas in us-west-2, and configure a scheduled AWS Backup job to replicate DB snapshots every 4 hours to meet the RPO target.
  5. E
    Deploy Amazon Aurora PostgreSQL in us-east-1 and deploy RDS Multi-AZ replica standby instances in us-west-2 to serve read requests and scale dynamic workloads automatically during normal operation.

Answer

Deploy an Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2, and create a Route 53 Private Hosted Zone (PHZ) containing a CNAME record pointing to the active database endpoint, associating this PHZ with the VPCs in both us-east-1 and us-west-2.
To meet the RPO of 1 minute and RTO of 10 minutes, deploying an Amazon Aurora Global Database is the most suitable strategy. It replicates data cross-region with latency typically under a second and allows quick promotion of the secondary cluster during a failover. Additionally, to allow client application servers in both regions to resolve a single database private DNS name, a Route 53 Private Hosted Zone must be associated with the VPCs in both us-east-1 and us-west-2.

Step-by-Step Solution

1
Analyze RTO and RPO requirements.
The RPO is 1 minute, which requires near real-time data replication between us-east-1 and us-west-2. The RTO is 10 minutes, which demands a fast failover mechanism.
Choosing the replication strategy depends on how much data loss (RPO) and downtime (RTO) the business can tolerate.
2
Select the database replication topology.
Amazon Aurora Global Database provides physical, sub-second cross-region replication, which easily satisfies the 1-minute RPO. It also allows the secondary cluster to be promoted to primary in minutes, meeting the 10-minute RTO.
Standard snapshot replication fails the RPO target, and RDS Multi-AZ standby instances do not support cross-region failover or read workload scaling.
3
Configure DNS routing for private resolution across VPCs.
Create a single Route 53 Private Hosted Zone (PHZ) and associate it with both the primary VPC in us-east-1 and the secondary VPC in us-west-2.
Without associating the PHZ with the secondary VPC, the clients in us-west-2 will not be able to resolve the database DNS name privately.

Key Concept

Multi-region disaster recovery design using Amazon Aurora Global Database and Route 53 Private Hosted Zone cross-VPC association.
Rate this question