Which solution will meet these requirements with the LEAST operational overhead?
Copy data from the source database to Amazon S3 by using AWS DataSync. Configure AWS Lambda functions to copy the data to the target database. Configure Amazon CloudWatch alarms to monitor the Lambda functions for errors and throttles. Use an Amazon Simple Notification Service (Amazon SNS) topic for email notification.
Create Amazon CloudWatch alarms to monitor DMS replication task metrics and host metrics. Use an Amazon Simple Notification Service (Amazon SNS) topic for email notification and to invoke an AWS Lambda function to configure a standby DMS replication instance in a different AWS Region.
Create Amazon CloudWatch alarms to monitor DMS replication task metrics and host metrics. Use an Amazon Simple Notification Service (Amazon SNS) topic for email notification. After receiving the notification, configure a new DMS replication task in the same AWS Region.
Modify the DMS replication instance by tuming on Multi-AZ support. Create Amazon CloudWatch alarms to monitor DMS replication task metrics and host metrics. Use an Amazon Simple Notification Service (Amazon SNS) topic for email notification.
Explanations:
AWS DataSync and Lambda functions introduce additional complexity and operational overhead. This approach does not directly address the issue of replication instance resiliency or interruptions in DMS, and requires manual intervention to copy data.
While CloudWatch alarms and SNS notifications are useful, automatically invoking a Lambda function to create a new replication instance in a different region is complex and does not directly improve the resiliency of the DMS replication instance itself.
This option involves manual intervention to configure a new DMS replication task after receiving an SNS notification. It introduces operational overhead and does not fully automate the failover process for replication instance resiliency.
Enabling Multi-AZ support for the DMS replication instance improves its resiliency by providing automatic failover. CloudWatch alarms and SNS notifications allow the team to monitor the health of the replication task and be alerted in case of issues, with minimal operational overhead.