Examkin

Question

Difficulty: HardHigh Availability and Disaster Recovery Design

A financial services enterprise is building a new online banking platform deployed across two regions: `us-east-1` and `us-west-2`. The platform consists of a containerized application running on Amazon ECS Fargate and an Amazon Aurora PostgreSQL database. The business requirements specify a Recovery Time Objective (RTO) of 1515 minutes and a Recovery Point Objective (RPO) of 11 minute. The public API layer must serve users with the lowest network latency during normal operations, but automatically redirect users to the other region if a regional outage occurs. Additionally, the application VPCs in both regions need to resolve internal endpoints hosted in a Private Hosted Zone (PHZ) within a shared services VPC. Outbound payment gateway integration requires highly available outbound connectivity in each region. Which architecture design meets these requirements while minimizing single points of failure?

  1. Configure Route 53 with Latency-based routing records pointing to regional Application Load Balancers, with Route 53 health checks associated with each record. Deploy Application Load Balancers and ECS Fargate tasks across multiple Availability Zones in each region, deploying a NAT Gateway in each Availability Zone. Use Amazon Aurora Global Database for data replication, and associate the shared services Private Hosted Zone with the application VPCs in both regions.Answer
  2. B
    Configure Route 53 with Latency-based routing records pointing to regional Application Load Balancers, with Route 53 health checks associated with each record. Deploy Application Load Balancers and ECS Fargate tasks across multiple Availability Zones in each region, routing all outbound traffic through a single NAT Gateway in each region to minimize costs. Use Amazon Aurora Global Database for data replication, and associate the shared services Private Hosted Zone with the application VPCs in both regions.
  3. C
    Configure Route 53 with a Failover routing policy, setting the Application Load Balancer in `us-east-1` as primary and the Application Load Balancer in `us-west-2` as secondary. Deploy Application Load Balancers and ECS Fargate tasks across multiple Availability Zones in each region, deploying a NAT Gateway in each Availability Zone. Use Amazon Aurora Global Database for data replication, and associate the shared services Private Hosted Zone with the application VPCs in both regions.
  4. D
    Configure Route 53 with Latency-based routing records pointing to regional Application Load Balancers, with Route 53 health checks associated with each record. Deploy Application Load Balancers and ECS Fargate tasks across multiple Availability Zones in each region, deploying a NAT Gateway in each Availability Zone. Use Amazon Aurora Global Database for data replication, and keep the Private Hosted Zone associated only with the shared services VPC, relying on default VPC DNS resolution for the application VPCs.

Answer

The architecture utilizing Route 53 Latency-based routing with health checks, multi-AZ deployment with redundant NAT Gateways, Aurora Global Database, and Private Hosted Zone associations meets all criteria.
The configuration using Latency-based routing with health checks, multiple NAT Gateways, Aurora Global Database, and proper Private Hosted Zone association is correct because it ensures that public traffic is routed based on latency with automatic failover, maintains outbound high availability by deploying a NAT Gateway in every Availability Zone, meets the low RTO/RPO targets using Aurora Global Database replication, and enables internal DNS resolution across regional VPC boundaries by associating the Private Hosted Zone.

Step-by-Step Solution

1
Evaluate user routing requirements.
Using Route 53 Latency-based routing with health checks correctly ensures users route to the region with the lowest latency under normal operations and automatically redirect in the event of an outage.
Failover routing only directs traffic to a single active region, which violates the requirement for low-latency routing for all users.
2
Evaluate network path redundancy.
Deploying a NAT Gateway in each Availability Zone provides redundant paths for outbound traffic.
A single NAT Gateway per region is a single point of failure for Fargate tasks in other Availability Zones during an AZ outage.
3
Evaluate cross-VPC private DNS resolution.
Associating the Private Hosted Zone from the shared services VPC with the application VPCs in both regions allows regional Fargate tasks to resolve internal endpoints.
Private Hosted Zones cannot be resolved by default VPC DNS unless explicitly associated with the requesting VPCs.
4
Verify replication strategy and disaster recovery boundaries.
Amazon Aurora Global Database provides sub-second cross-region replication (meeting the 11 minute RPO) and managed failover within minutes (meeting the 1515 minutes RTO).
This strategy satisfies the strict recovery objectives requested by the business scenario.

Key Concept

Designing multi-region disaster recovery setups requires combining latency-sensitive routing with redundant local networking infrastructure, database global replication, and cross-VPC private DNS resource sharing.
Rate this question

Topics

High Availability and Disaster Recovery Design

More questions from this topic