Examkin

Soru

Zorluk: ZorResilient Database Configurations and High Availability

An enterprise is migrating a critical, write-heavy core banking application to AWS. The database architecture must satisfy the following resiliency requirements:
- Within the primary AWS region, the database must achieve high availability with zero data loss (Recovery Point Objective, RPO=0\text{RPO} = 0) and automated failover in less than 30 seconds.
- Across a secondary AWS region, the disaster recovery solution must support active read scaling during normal operations, maintain a cross-region replication lag under 2 seconds (RPO<2\text{RPO} < 2 seconds), and allow failover to the secondary region in less than 5 minutes (RTO<5\text{RTO} < 5 minutes).

Which database configuration meets these requirements with the lowest operational overhead?

  1. Amazon Aurora PostgreSQL-Compatible Edition with an Aurora Global Database. Deploy the primary cluster in the main region and a secondary cluster in the recovery region, routing read traffic in the secondary region to the local reader instances.Cevap
  2. B
    Amazon RDS for PostgreSQL with a Multi-AZ DB instance deployment in the primary region, and a cross-region Read Replica in the secondary region. Use Route 53 Latency Routing to automatically direct database write traffic to the replica in the secondary region if the primary region becomes unavailable.
  3. C
    Amazon RDS for PostgreSQL with a Multi-AZ DB instance deployment in the primary region. In the secondary region, configure a pilot light database environment by restoring daily snapshots transferred from the primary region, scale up the instance class only during failover, and direct traffic using Route 53 Failover Routing.
  4. D
    Amazon RDS for PostgreSQL with a Single-AZ DB instance deployment in the primary region. Configure an asynchronous cross-region Read Replica in the secondary region, and write a custom AWS Lambda script to monitor primary database health and automatically promote the replica during a regional outage.

Cevap

Amazon Aurora PostgreSQL-Compatible Edition with an Aurora Global Database. Deploy the primary cluster in the main region and a secondary cluster in the recovery region, routing read traffic in the secondary region to the local reader instances.
The correct option implements Amazon Aurora Global Database, which matches all constraints. For local high availability, Aurora replicates data synchronously across three AZs, ensuring RPO=0\text{RPO} = 0 and automated failover in less than 30 seconds. For multi-region resilience, storage-based physical replication keeps lag below 2 seconds (typically under 1 second), and secondary clusters can serve active read traffic and be promoted to primary in less than a minute.

Adım Adım Çözüm

1
Analyze single-region high availability requirements.
The solution requires RPO=0\text{RPO} = 0 (zero data loss) and automated failover in less than 30 seconds within the primary region. Aurora replicates data across three AZs with shared storage, achieving synchronous-like durabilities and automatic failover within 30 seconds, meeting the local HA requirements.
Ensures the database survives an Availability Zone outage without data loss and with minimal recovery time.
2
Analyze cross-region disaster recovery and read-scaling requirements.
The secondary region needs read scaling, RPO<2\text{RPO} < 2 seconds, and RTO<5\text{RTO} < 5 minutes. Aurora Global Databases use storage-based replication with a typical lag of less than 1 second, allow reader instances in the secondary region to serve local read traffic, and can be promoted to primary within 1 minute.
Guarantees that regional disaster recovery requirements are met while optimizing read performance for secondary region users.
3
Evaluate the operational overhead and suitability of alternative services.
RDS PostgreSQL Multi-AZ with cross-region read replicas requires manual intervention or complex automation for cross-region failover and does not support storage-level global replication. Restoring from daily snapshots (pilot light) violates the 2-second RPO requirement.
Selects the option that minimizes custom automation and meets all RPO/RTO constraints natively.

Anahtar Kavram

Amazon Aurora Global Databases and Multi-AZ replication
Bu soruyu puanla

Konular

Resilient Database Configurations and High Availability

Bu konudan daha fazla soru