Which solution will meet these requirements with the MOST operational efficiency?
Create an Aurora read replica in us-west-1 similar in size to the production application’s Aurora MySQL cluster writer instance.
Convert the Aurora cluster to an Aurora global database. Configure managed failover.
Create a new Aurora cluster in us-west-1 that has Cross-Region Replication.
Create a new Aurora cluster in us-west-1. Use AWS Database Migration Service (AWS DMS) to sync both clusters.
Explanations:
Creating an Aurora read replica in us-west-1 would not meet the RPO of 5 minutes, as read replicas typically lag behind the primary instance. Additionally, failover to a read replica requires manual intervention, which may not meet the RTO of 20 minutes.
Converting the Aurora cluster to an Aurora global database allows for near real-time replication between regions, meeting the 5-minute RPO. It also supports automatic failover, enabling a recovery within the 20-minute RTO with minimal configuration changes.
Creating a new Aurora cluster in us-west-1 with Cross-Region Replication would not ensure the 5-minute RPO since replication can experience delays. Furthermore, it requires additional setup and management for failover, impacting operational efficiency.
Using AWS Database Migration Service (AWS DMS) to sync both clusters adds complexity and operational overhead. It may not consistently meet the 5-minute RPO, and the RTO may not be achievable within 20 minutes due to the need for manual intervention and setup during failover.
I sort that the answer is:
Convert the Aurora cluster to an Aurora global database. Configure managed failover.