AWS Certified DevOps Engineer – ProfessionalResilient Cloud SolutionsHard
A financial services company is migrating its critical transaction processing application to AWS. The application requires extremely low latency and high availability across multiple Availability Zones. Data consistency is paramount, and the company needs to ensure that in the event of a regional disaster, recovery time objectives (RTO) and recovery point objectives (RPO) are minimized to less than 15 minutes. The current on-premises setup uses a synchronous replication database. Which AWS database and replication strategy best meets these requirements?
- AAmazon DynamoDB Global Tables with strongly consistent reads enabled.
- BAmazon RDS for PostgreSQL with Multi-AZ deployment and logical replication to a standby instance in another region.
- CAmazon Aurora PostgreSQL-Compatible Edition with Multi-AZ DB cluster deployment and Aurora Global Database.
- DAmazon EC2 instances with self-managed PostgreSQL and streaming replication to a cross-region standby.
Show answer & explanationAnswer & explanation
Correct answer: C. Amazon Aurora PostgreSQL-Compatible Edition with Multi-AZ DB cluster deployment and Aurora Global Database.
Aurora Global Database provides fast, low-latency cross-region replication for disaster recovery with RTO/RPO typically measured in minutes. Multi-AZ DB cluster ensures high availability within a region and automatic failover.
Why the other options are wrong
- A. DynamoDB Global Tables are eventually consistent by default for cross-region replication. While strong consistency can be enabled for reads within a region, the cross-region replication itself is not synchronously consistent for all writes, and it's a NoSQL database, which might not fit 'transaction processing' if relational features are needed.
- B. Logical replication for RDS PostgreSQL is asynchronous and typically has higher RTO/RPO than required for cross-region disaster recovery.
- D. Self-managed EC2 with streaming replication requires significant operational overhead, and achieving RTO/RPO of less than 15 minutes reliably across regions with synchronous replication can be complex and expensive to manage without Aurora's built-in features.
Aurora Global Database
Amazon Aurora Global Database allows a single Amazon Aurora database to span multiple AWS regions, enabling fast disaster recovery from regional outages and low-latency global reads.
- Cross-region replication for disaster recovery.
- RPO typically in seconds, RTO in minutes.
- Secondary regions are read-only, can be promoted to primary.
- Uses dedicated infrastructure for replication, minimizing performance impact on primary.
Memory trick: Global-scale resilience needs a global database shield.