Question

Difficulty: MediumHigh Availability and Disaster Recovery Design

An educational technology company is designing a new online testing platform that will serve students globally. The application's web tier runs on Amazon EC2 instances inside private subnets across two Availability Zones in the primary region (us-east-1). Outbound internet access for grading APIs is routed through NAT Gateways. The database is hosted on Amazon Aurora PostgreSQL. The architecture must support a secondary recovery region (us-west-2) to satisfy a disaster recovery plan with a Recovery Point Objective (RPO) of 2 minutes2\text{ minutes} and a Recovery Time Objective (RTO) of 15 minutes15\text{ minutes}. For internal service communications, a Route 53 Private Hosted Zone (PHZ) is used to resolve internal microservice endpoints. Which of the following architectures should a solutions architect recommend to meet the RTO and RPO requirements while ensuring high availability and fault tolerance?

  1. A
    Deploy the EC2 web tier in both us-east-1 and us-west-2. Deploy a single NAT Gateway in us-east-1 and route outbound traffic from both Availability Zones through it. Use Amazon Aurora Global Database to replicate the database from us-east-1 to us-west-2. Associate the Route 53 Private Hosted Zone (PHZ) with the VPCs in both regions. Configure Route 53 failover routing policies with active-passive health checks to route public traffic.
  2. B
    Deploy the EC2 web tier in both us-east-1 and us-west-2. Deploy a NAT Gateway in each Availability Zone within each region. Use Amazon Aurora Global Database to replicate the database from us-east-1 to us-west-2. Associate the Route 53 Private Hosted Zone (PHZ) only with the us-east-1 VPC. Configure Route 53 failover routing policies with active-passive health checks to route public traffic.
  3. Deploy the EC2 web tier in both us-east-1 and us-west-2. Deploy a NAT Gateway in each Availability Zone within each region. Use Amazon Aurora Global Database to replicate the database from us-east-1 to us-west-2. Associate the Route 53 Private Hosted Zone (PHZ) with the VPCs in both regions. Configure Route 53 failover routing policies with active-passive health checks to route public traffic to us-east-1, failing over to us-west-2 if the primary region is degraded.Answer
  4. D
    Deploy the EC2 web tier in both us-east-1 and us-west-2. Deploy a NAT Gateway in each Availability Zone within each region. Implement a backup schedule using AWS Backup to create snapshots of the Aurora database in us-east-1 every 12 hours12\text{ hours} and copy them to us-west-2. Associate the Route 53 Private Hosted Zone (PHZ) with the VPCs in both regions. Configure Route 53 failover routing policies to route public traffic, and restore the database from the copied snapshot in us-west-2 during a failover event.

Answer

The correct architecture replicates the database using Amazon Aurora Global Database, deploys a NAT Gateway per Availability Zone in each region for high availability, associates the Route 53 Private Hosted Zone with VPCs in both regions, and uses Route 53 failover routing policies with active-passive health checks to route public traffic.
The correct architecture replicates the database using Amazon Aurora Global Database, which offers sub-second data replication and quick failover capabilities, satisfying both the 2 minutes2\text{ minutes} RPO and 15 minutes15\text{ minutes} RTO. Deploying a NAT Gateway in each Availability Zone in both regions prevents outbound connectivity failures. Associating the Private Hosted Zone with the VPCs in both regions ensures internal DNS resolution remains functional after failover. Finally, active-passive Route 53 failover routing policies ensure traffic is dynamically redirected to the secondary region if the primary region goes down.

Step-by-Step Solution

1
Evaluate the database replication strategy against RTO and RPO requirements.
Amazon Aurora Global Database provides sub-second replication lag (meeting the RPO of 2 minutes2\text{ minutes}) and managed failovers that take under 2 minutes2\text{ minutes} (meeting the RTO of 15 minutes15\text{ minutes}). Snapshot-based recovery is too slow and violates RPO.
Choosing the correct replication technology is critical to achieving near-zero recovery objectives.
2
Assess the routing and network redundancy requirements.
To avoid a single point of failure for outbound traffic, a NAT Gateway must be deployed in each Availability Zone of the VPC. A single NAT Gateway configuration is not highly available.
High availability requires removing single points of failure across all architectural tiers.
3
Configure internal DNS resolution and external failover routing.
The Route 53 Private Hosted Zone (PHZ) must be associated with the VPCs in both the primary and secondary regions to resolve internal endpoints. Public traffic failover must be handled using Route 53 active-passive failover routing policies with health checks.
DNS resolution must function correctly in both regions to support application failover.

Key Concept

Multi-region disaster recovery and high availability architecture design
Rate this question