Examkin

Question

Difficulty: Very hardHigh Availability and Disaster Recovery Design

A healthcare provider is designing a SaaS clinical trials management application deployed across two AWS Regions: `us-east-1` (primary) and `us-west-2` (secondary). The application uses Amazon ECS on AWS Fargate for compute and an Amazon Aurora PostgreSQL database. The architecture must achieve a recovery point objective (RPO) of 1 minute and a recovery time objective (RTO) of 15 minutes for disaster recovery. Client requests must be routed to the primary region during normal operations, and automatically failover to the secondary region if the primary region becomes unhealthy. Internally, microservices in different VPCs across both regions must resolve the database's private endpoints using a shared Route 53 Private Hosted Zone (PHZ). Outbound internet connectivity from Fargate tasks to external health registries must be resilient against Availability Zone (AZ) outages in each region.

Which two options should the Solutions Architect implement to satisfy these requirements?

  1. Configure Route 53 public DNS with a primary failover routing policy pointing to the Application Load Balancer (ALB) in `us-east-1` and a secondary failover routing policy pointing to the ALB in `us-west-2`, linking both to health checks. Create an Amazon Aurora Global Database with the primary cluster in `us-east-1` and a secondary cluster in `us-west-2`. Associate the shared Route 53 PHZ containing the database endpoint with all VPCs in both regions.Answer
  2. Deploy a NAT Gateway in each Availability Zone of the VPCs in both regions, and configure the private subnets' route tables to point outbound internet traffic to the local NAT Gateway in their respective Availability Zone.Answer
  3. C
    Deploy a single NAT Gateway in the public subnet of the primary Availability Zone in each region, and point the route tables of all private subnets across all Availability Zones to this NAT Gateway to minimize monthly NAT gateway deployment costs.
  4. D
    Configure Route 53 public DNS with latency-based routing policies pointing to the Application Load Balancer in each region. Configure Aurora cross-region read replicas, and use AWS Backup to perform daily snapshots of the database. During a disaster recovery event, promote the replica and restore from the latest snapshot if data corruption occurs.
  5. E
    Create an Amazon Aurora Global Database with the primary cluster in `us-east-1` and a secondary cluster in `us-west-2`. Create the Route 53 PHZ only in the primary region's VPC, and configure a Route 53 Resolver outbound endpoint in the primary VPC to forward DNS queries from the secondary region's VPCs without associating the PHZ to the secondary region's VPCs.

Answer

To meet the requirements, the Solutions Architect must implement Route 53 public DNS failover routing, Amazon Aurora Global Database with the Private Hosted Zone associated across all VPCs in both regions, and deploy redundant NAT Gateways in each Availability Zone in both regions.
The correct solution requires configuring Route 53 failover routing to direct public traffic to the active region and failover when needed, utilizing Aurora Global Database to achieve low latency replication (seconds RPO) and rapid failover (seconds/minutes RTO) to meet the 1-minute RPO and 15-minute RTO targets. Additionally, Route 53 Private Hosted Zones must be associated with each VPC in both regions to allow internal microservices to resolve the database private endpoints. To ensure high availability of outbound connections for the Fargate tasks, NAT Gateways must be deployed in each Availability Zone so that an AZ outage does not isolate the tasks running in the other AZs.

Step-by-Step Solution

1
Set up Route 53 public DNS failover records pointing to the ALBs in the primary and secondary regions, linked to health checks.
Establishes automatic client-side regional failover (active-passive routing).
Allows traffic to be redirected from the primary region to the secondary region if the primary region goes down.
2
Provision an Amazon Aurora Global Database with the primary cluster in the primary region and the secondary cluster in the secondary region.
Enables physical cross-region storage replication with sub-minute latency.
Supports the 1-minute RPO and 15-minute RTO disaster recovery requirements.
3
Associate the Route 53 Private Hosted Zone (PHZ) containing the database endpoint with all VPCs across both regions.
Allows microservices in all VPCs to resolve the database private domain name locally.
Correctly configures private DNS resolution across VPCs and regions without relying on invalid forwarding loops.
4
Provision a NAT Gateway in each public subnet across all Availability Zones in both regions.
Provides redundant outbound paths for ECS Fargate tasks in the private subnets.
Protects against Availability Zone outages, preventing an outage in one AZ from breaking outbound internet access for resources in other AZs.

Key Concept

Designing multi-region active-passive architectures with Aurora Global Database, Private Hosted Zone VPC association, and multi-AZ NAT Gateway redundancy.
Rate this question

Topics

High Availability and Disaster Recovery Design

More questions from this topic