Examkin

Soru

Zorluk: ZorHigh Availability and Disaster Recovery Design

A financial technology startup is designing a new multi-region transaction processing application. The primary infrastructure is located in the `us-east-1` Region, consisting of an Amazon ECS cluster running on Amazon EC2 instances in an Auto Scaling group (ASG) behind an Application Load Balancer (ALB). The database tier uses an Amazon Aurora PostgreSQL cluster. For internal service discovery, microservices resolve endpoints using a Route 53 Private Hosted Zone named `fintech.internal`.

The startup must establish a disaster recovery (DR) site in the `us-west-2` Region to comply with regulatory standards. The architecture must satisfy a Recovery Time Objective (RTO) of 1515 minutes and a Recovery Point Objective (RPO) of 11 minute. Outbound internet access from the application instances is required in both regions to communicate with external payment gateways, and this outbound path must survive the failure of any single Availability Zone (AZ) within a region. During normal operations, all public customer traffic must be routed to `us-east-1`, and costs in the secondary region must be minimized.

Which of the following architectures meets these requirements with the lowest cost and operational overhead?

  1. Configure Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2. In us-west-2, deploy the ALB and configure the ECS ASG with a minimum capacity of 11 instance. In both VPCs, deploy a NAT Gateway in each Availability Zone and associate private subnet route tables with the local NAT Gateway in the same Availability Zone. Associate the fintech.internal Private Hosted Zone with both VPCs. Create Route 53 failover routing records in the public hosted zone, setting the us-east-1 ALB as primary with health checks, and the us-west-2 ALB as secondary.Cevap
  2. B
    Configure daily cross-region copies of Amazon Aurora database snapshots from us-east-1 to us-west-2, and restore the database cluster in us-west-2 during failover. In us-west-2, deploy the ALB and configure the ECS ASG with a minimum capacity of 00 instances. In both VPCs, deploy a NAT Gateway in each Availability Zone. Associate the fintech.internal Private Hosted Zone with both VPCs. Create Route 53 failover routing records in the public hosted zone pointing to the respective ALBs.
  3. C
    Configure Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2. In us-west-2, deploy the ALB and configure the ECS ASG with a minimum capacity of 11 instance. Deploy a single NAT Gateway in us-west-2 across a single Availability Zone, and update the route tables of all private subnets in the region to route internet-bound traffic through this NAT Gateway. Associate the fintech.internal Private Hosted Zone with both VPCs. Create Route 53 failover routing records in the public hosted zone pointing to the respective ALBs.
  4. D
    Configure Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2. In us-west-2, deploy the ALB and configure the ECS ASG with a minimum capacity of 11 instance. Deploy a NAT Gateway in each Availability Zone in both regions. Associate the fintech.internal Private Hosted Zone with the us-east-1 VPC. Create Route 53 latency routing records in the public hosted zone pointing to both the us-east-1 ALB and us-west-2 ALB.

Cevap

Configure Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2. In us-west-2, deploy the ALB and configure the ECS ASG with a minimum capacity of 1 instance. In both VPCs, deploy a NAT Gateway in each Availability Zone and associate private subnet route tables with the local NAT Gateway in the same Availability Zone. Associate the fintech.internal Private Hosted Zone with both VPCs. Create Route 53 failover routing records in the public hosted zone, setting the us-east-1 ALB as primary with health checks, and the us-west-2 ALB as secondary.
The correct architecture uses Amazon Aurora Global Database to achieve near-zero replication lag, satisfying the 1-minute RPO requirement. An ECS Auto Scaling group with a minimum capacity of 1 instance in the recovery region ensures that the standby ALB has at least one active backend target, allowing the Route 53 health check to pass while minimizing compute costs during normal operations. Outbound NAT gateway redundancy (deploying a NAT Gateway in each Availability Zone) ensures that internet-bound traffic remains operational if a single Availability Zone experiences an outage. Associating the internal Route 53 Private Hosted Zone with both VPCs ensures that microservices can resolve internal endpoints in either region. Finally, Route 53 Failover routing records automatically redirect public traffic from the primary region to the secondary region if the primary load balancer fails health checks.

Adım Adım Çözüm

1
Select a database replication strategy that satisfies the RPO requirement.
Amazon Aurora Global Database provides sub-second replication lag to the secondary region, meeting the 11-minute RPO constraint, whereas daily snapshots only offer a 2424-hour RPO.
Regulatory compliance demands near-real-time replication to prevent transaction data loss in a disaster scenario.
2
Ensure outbound connectivity meets the high availability criteria in both regions.
Deploying one NAT Gateway per Availability Zone (AZ) in both regions ensures that a failure in a single AZ does not disrupt outbound internet traffic for resources in other AZs.
A single regional NAT Gateway is a single point of failure, violating the requirement that outbound paths survive any single AZ outage.
3
Establish cross-region DNS resolution and failover configurations.
Associate the Route 53 Private Hosted Zone with both the primary and standby VPCs, and create Route 53 failover routing records pointing to the ALBs.
Private Hosted Zones must be explicitly associated with each VPC that requires internal domain resolution. Failover routing ensures all public traffic stays in the primary region during normal operations.
4
Configure the standby application tier to support health checking while minimizing cost.
Set the ECS Auto Scaling Group in the secondary region to a minimum capacity of 11 running instance behind the secondary Application Load Balancer.
If capacity is set to 00, the secondary ALB health check will fail, which prevents Route 53 from routing traffic to it during a failover event. A Warm Standby with 1 instance minimizes cost while remaining functional.

Anahtar Kavram

Multi-Region High Availability and Disaster Recovery Design with Aurora Global Database, Route 53 failover, NAT Gateway redundancy, and Private Hosted Zone sharing.
Bu soruyu puanla

Konular

High Availability and Disaster Recovery Design

Bu konudan daha fazla soru