Examkin

Soru

Zorluk: Çok zorEnhancing Reliability and Disaster Recovery

An enterprise runs a high-volume payment processing application in a single AWS account in the us-east-1 Region. The system uses an Amazon Elastic Kubernetes Service (EKS) cluster for compute and an Amazon DynamoDB table for transaction states. The application has suffered outages due to regional service issues and internet routing degradation. To improve availability, the solutions architect must redesign the architecture to implement a multi-region Active-Passive disaster recovery (DR) strategy across us-east-1 and us-west-2. The solution must achieve a Recovery Time Objective (RTO) of less than 5 minutes and a Recovery Point Objective (RPO) of less than 1 minute. Which of the following actions should the solutions architect take to meet these reliability and disaster recovery requirements? (Select TWO.)

  1. Convert the existing DynamoDB table to a global table by adding a replica in the us-west-2 Region, and provision Route 53 Application Recovery Controller (ARC) routing controls linked to Route 53 failover routing records pointing to the Application Load Balancers in each region.Cevap
  2. Deploy a duplicate EKS cluster in the us-west-2 Region with pre-provisioned worker nodes matching the baseline capacity of the primary region, and ensure the application deployment pipeline deploys container images and configuration maps to both EKS clusters to maintain configuration alignment.Cevap
  3. C
    Configure a pilot light DR strategy by maintaining a backup EKS cluster in the us-west-2 Region with an Auto Scaling Group size set to zero, and use an AWS Lambda function triggered by Route 53 DNS failover events to bootstrap the EKS nodes and pull the container images when an outage is detected in the us-east-1 Region.
  4. D
    Set up Route 53 active-passive failover using standard Route 53 HTTP/HTTPS health checks directly targeting the Application Load Balancers in both regions to automate DNS failover without manual intervention during a regional control plane event.
  5. E
    Establish an AWS Transit Gateway in both regions to route all outbound egress payment traffic from the us-west-2 EKS cluster through a single NAT Gateway in the us-east-1 Region to preserve static IP addresses for external payment providers.

Cevap

Convert the existing DynamoDB table to a global table, provision Route 53 ARC routing controls, deploy a duplicate EKS cluster in the secondary region with pre-provisioned baseline worker nodes, and keep application configurations synchronized across both clusters.
To achieve an RPO of under 1 minute and an RTO of under 5 minutes, database replication must be continuous and asynchronous, which is fulfilled by DynamoDB Global Tables. Furthermore, the standby EKS cluster must be pre-provisioned to run containerized workloads immediately without cold-start bootstrapping latency. Traffic routing must be managed deterministically using Route 53 Application Recovery Controller (ARC) routing controls, which bypass control plane dependencies of the failing region.

Adım Adım Çözüm

1
Address the data replication requirements to meet the sub-minute recovery point objective (RPO).
Convert the DynamoDB table to a Global Table with replicas in both the primary and standby regions, enabling asynchronous replication with sub-second latency.
This guarantees that database transaction states are continuously replicated and available in the disaster recovery region with an RPO well below the 1-minute limit.
2
Ensure the compute infrastructure is pre-warmed to meet the strict 5-minute recovery time objective (RTO).
Deploy the secondary EKS cluster with pre-provisioned node capacity matching baseline requirements and configure the CI/CD pipeline to deploy container images and ConfigMaps to both regions.
Cold startup of EKS nodes, Kubernetes agent initialization, and pulling heavy container images during a failover event would exceed the 5-minute RTO.
3
Implement a highly reliable failover mechanism that does not depend on the control plane of the degraded region.
Create Route 53 Application Recovery Controller (ARC) routing controls and associate them with Route 53 failover records.
Route 53 ARC uses an extremely resilient, globally distributed data plane for state updates, allowing failover execution within seconds even during a full regional control plane failure.

Anahtar Kavram

Designing multi-region disaster recovery for applications requiring low RTO and RPO requires active data replication (such as DynamoDB Global Tables) and pre-provisioned, hot/warm standby compute resources combined with deterministic routing failover using Route 53 ARC.
Bu soruyu puanla

Konular

Enhancing Reliability and Disaster Recovery