AWS Certified DevOps Engineer – Professional (DOP-C02) — Question 390
A company uses Amazon RDS for Microsoft SQL Server as its primary database for applications. The company needs to ensure high availability within and across AWS Regions.
An Amazon Route 53 CNAME record is configured for the database endpoint. The applications connect to the database endpoint. The company must redirect application traffic to a standby database during a failover event. The company must maintain an RPO of less than 1 minute and an RTO of less than 10 minutes.
Which solution will meet these requirements?
Answer options
- A. Deploy an Amazon RDS for SQL Server Multi-AZ DB cluster deployment that uses cross-Region read replicas. Use automation to promote the read replica to a standalone instance and to update the Route 53 record.
- B. Deploy an Amazon RDS for SQL Server Multi-AZ DB cluster deployment. Set up automated snapshots to be copied to another Region every 5 minutes. Use AWS Lambda to restore the latest snapshot in the secondary Region during failover.
- C. Deploy an Amazon RDS for SQL Server Single-AZ DB instance. Use AWS Database Migration Service (AWS DMS) to replicate data continuously to an RDS DB instance in another Region. Use Amazon CloudWatch alarms to notify the company about failover events.
- D. Deploy an Amazon RDS for SQL Server Single-AZ DB instance. Configure AWS Backup to create cross-Region backups every 30 seconds. Use automation to restore the latest backup and to update the Route 53 record during failover.
Correct answer: A
Explanation
Option A is correct because a Multi-AZ DB cluster deployment provides high availability within the primary Region, while the cross-Region read replica ensures data is continuously replicated to another Region with sub-minute latency (meeting the < 1 minute RPO). Promoting a read replica to a primary database and updating Route 53 can be completed in a few minutes, satisfying the < 10 minutes RTO. Options B and D fail the RTO requirement because restoring a database from a snapshot or backup takes longer than 10 minutes, and Option C uses a Single-AZ instance which does not provide high availability within the primary Region.