Soru

Zorluk: OrtaHigh Availability and Disaster Recovery Design

A healthcare technology company is building a patient telemetry monitoring system on AWS. The architecture consists of an application tier running on Amazon EC2 instances in an Auto Scaling group across multiple Availability Zones, and a database tier running on an Amazon Aurora PostgreSQL-Compatible Edition cluster. The system must ingest real-time telemetry data from local hospital gateways. The company requires a high-availability design in the primary region of us-east-1, and a disaster recovery (DR) setup in us-west-2 that achieves a Recovery Time Objective (RTO) of 15 minutes and a Recovery Point Objective (RPO) of 1 minute. The EC2 instances must be able to securely access external APIs on the internet to retrieve patient metadata. Additionally, the application tier uses a Route 53 Private Hosted Zone for internal database endpoint resolution. Which configuration meets these requirements while minimizing cost and operational complexity?

  1. A
    Deploy the EC2 Auto Scaling group in us-east-1 across three Availability Zones (AZs) with a NAT Gateway in each AZ. Configure hourly snapshots of the Amazon Aurora cluster, copy them to us-west-2, and configure a script to restore the database from the latest snapshot in us-west-2 during a failover event. Configure a Route 53 Failover routing policy with health checks pointing to the public ALBs in both regions.
  2. Deploy the EC2 Auto Scaling group in us-east-1 across three Availability Zones (AZs) with a NAT Gateway in each AZ. Implement Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster with a read replica in us-west-2. Configure a Route 53 Failover routing policy with health checks pointing to the public Application Load Balancers (ALBs) in both regions, and associate the internal Route 53 Private Hosted Zone with both the us-east-1 and us-west-2 VPCs.Cevap
  3. C
    Deploy the EC2 Auto Scaling group in us-east-1 across three Availability Zones (AZs) with a single NAT Gateway in one of the AZs. Implement Amazon Aurora Global Database with the primary cluster in us-east-1 and a secondary cluster in us-west-2. Configure a Route 53 Latency routing policy with health checks pointing to the public ALBs in both regions, and associate the internal Route 53 Private Hosted Zone only with the us-east-1 VPC.
  4. D
    Deploy the EC2 Auto Scaling group in us-east-1 across three Availability Zones (AZs) with a NAT Gateway in each AZ. Set up a standard Amazon Aurora PostgreSQL cluster in us-east-1 and a separate cluster in us-west-2. Configure Aurora Replica Auto Scaling to automatically replicate data across regions and promote the us-west-2 cluster in the event of an outage in us-east-1. Configure a Route 53 Failover routing policy with health checks pointing to the public ALBs.

Cevap

The correct configuration deploys the EC2 Auto Scaling group in us-east-1 across three Availability Zones with a NAT Gateway in each AZ, implements Amazon Aurora Global Database with a secondary cluster in the failover region, configures a Route 53 Failover routing policy with health checks pointing to the public ALBs, and associates the internal Route 53 Private Hosted Zone with the VPCs in both regions.
The correct configuration uses a redundant multi-AZ networking setup in the primary region (with a NAT Gateway in each AZ to avoid a single point of failure) and implements Amazon Aurora Global Database for sub-second cross-region replication to meet the 1-minute RPO and 15-minute RTO. A Route 53 Failover routing policy uses health checks to automatically direct public ingress traffic to the secondary region. Finally, associating the internal Route 53 Private Hosted Zone with both regional VPCs ensures internal DNS name resolution continues to work seamlessly after failover.

Adım Adım Çözüm

1
Evaluate the database tier against the RTO of 15 minutes and RPO of 1 minute.
Amazon Aurora Global Database provides sub-second replication (meeting the 1-minute RPO) and can be promoted to a primary writer cluster in less than a minute (meeting the 15-minute RTO). Traditional cross-region snapshot backup and restore violates both RPO (limited by snapshot frequency) and RTO (restore time is too slow).
This rules out options that rely on backup-and-restore or standard configurations that do not support cross-region failover automation.
2
Assess the high-availability networking requirement for external API access in the primary region.
To ensure that outbound connectivity does not have a single point of failure, a NAT Gateway must be deployed in each Availability Zone where the application EC2 instances reside.
This eliminates configurations using a single NAT Gateway across multiple Availability Zones, which would lead to an outage if that specific zone failed.
3
Determine the DNS routing and Private Hosted Zone (PHZ) association requirements.
Route 53 Failover routing correctly sends public client traffic to the secondary region if the primary region goes down. To ensure internal name resolution works for application components in us-west-2 during a failover, the Route 53 PHZ must be associated with the VPCs in both regions.
Without cross-VPC association, the application servers in the secondary region would be unable to resolve internal database endpoints.

Anahtar Kavram

Designing a highly available multi-AZ architecture with cross-region active-passive disaster recovery using Route 53 failover routing, Aurora Global Database, and cross-VPC private hosted zone associations.
Tahmini Süre:2m 0s
Bu soruyu puanla