Examkin

Question

Difficulty: HardHigh Availability and Disaster Recovery Design

An international airline is designing a high-availability and disaster recovery architecture for its global crew scheduling and flight dispatching system. The system runs on Amazon EC2 instances in an Auto Scaling group behind an Application Load Balancer (ALB). The primary workload is hosted in `us-east-1`, and the disaster recovery region is `us-west-2`. The airline has defined a Recovery Time Objective (RTO) of 1010 minutes and a Recovery Point Objective (RPO) of 11 minute. The flight dispatchers are distributed globally, and their client connections must be routed to the region offering the lowest network latency. Furthermore, the application requires outbound internet access to retrieve real-time weather feeds from external aviation authorities, which must remain functional even if a single Availability Zone within either region suffers an outage. Which solution should a solutions architect recommend 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`. In both regions, deploy the EC2 instances across multiple Availability Zones with Route 53 latency-based routing records and health checks pointing to the regional ALBs. Run the application tier in `us-west-2` at minimum capacity and use an Auto Scaling group to scale it up during a failover. Configure NAT Gateways in each Availability Zone in both VPCs for outbound internet access.Answer
  2. B
    Deploy an Amazon Aurora Global Database with the primary cluster in `us-east-1` and a secondary cluster in `us-west-2`. In both regions, deploy the EC2 instances across multiple Availability Zones with Route 53 latency-based routing records and health checks pointing to the regional ALBs. Run the application tier in `us-west-2` at minimum capacity and use an Auto Scaling group to scale it up during a failover. To optimize resource costs, deploy a single NAT Gateway in the public subnet of the first Availability Zone in each VPC to handle outbound internet access.
  3. C
    Deploy an Amazon Aurora PostgreSQL DB cluster in `us-east-1` and configure AWS Backup to take automated snapshots every hour, copying them to `us-west-2`. In both regions, deploy the EC2 instances across multiple Availability Zones with Route 53 latency-based routing records and health checks pointing to the regional ALBs. Run the application tier in `us-west-2` at minimum capacity and use an Auto Scaling group to scale it up during a failover. Configure NAT Gateways in each Availability Zone in both VPCs for outbound internet access.
  4. D
    Deploy a Multi-AZ Amazon RDS PostgreSQL DB instance in `us-east-1`. To handle local read requests and reduce database load during failover, configure the application servers in `us-west-2` to read directly from the passive RDS standby instance in the secondary Availability Zone. Deploy the EC2 instances across multiple Availability Zones with Route 53 latency-based routing records and health checks pointing to the regional ALBs. Configure NAT Gateways in each Availability Zone in both VPCs for outbound internet access.

Answer

Deploy an Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2. Route client connections using Route 53 latency-based routing records with health checks pointing to regional ALBs. Run the application tier in us-west-2 at minimum capacity and use an Auto Scaling group to scale it up during failover. Deploy NAT Gateways in each Availability Zone in both VPCs for redundant outbound internet access.
The correct solution addresses all architecture requirements. First, Amazon Aurora Global Database replicates data across regions in under 11 second, which easily satisfies the 11-minute RPO. Second, Route 53 latency-based routing records with associated health checks ensure that clients are routed to the lowest-latency healthy endpoint, automatically failing over to the secondary region if the primary region goes down. Third, the application tier in the secondary region runs at a lower capacity to minimize idle costs and scales up using Auto Scaling during a failover to meet the 1010-minute RTO. Finally, deploying NAT Gateways in each Availability Zone of the subnets ensures that outbound connectivity remains operational even during an Availability Zone outage.

Step-by-Step Solution

1
Select a cross-region database replication mechanism that meets the RPO.
Amazon Aurora Global Database is chosen because it offers physical replication lag of less than 11 second across regions, satisfying the 11-minute RPO constraint.
Standard backup/restore or read replica copying mechanisms cannot guarantee an RPO under 11 minute, and RDS Multi-AZ standby instances do not span regions.
2
Design the DNS routing policy to handle low-latency and automatic failover.
Amazon Route 53 latency-based routing is configured with health checks associated with the primary and secondary region Application Load Balancers.
This configuration routes users to the closest region under normal operations and automatically redirects traffic to the surviving region if a failure occurs, satisfying both low latency and the 1010-minute RTO.
3
Ensure outbound connectivity is highly available and resilient to AZ outages.
NAT Gateways are deployed in each Availability Zone in both regions where the application instances reside.
If a single NAT Gateway were shared across multiple Availability Zones in a VPC, the outage of the zone hosting that NAT Gateway would disrupt outbound connectivity for the entire region. Multi-AZ NAT Gateway deployment removes this single point of failure.

Key Concept

Multi-region disaster recovery design balancing RTO/RPO using Aurora Global Database, Route 53 latency routing with health checks, and AZ-redundant NAT Gateways.
Rate this question

Topics

High Availability and Disaster Recovery Design

More questions from this topic