An application runs behind an Application Load Balancer (ALB) in the us-west-2 Region. A Route 53 record set contains an alias for app.anycompany.com that points to that ALB using a simple routing policy. The company has deployed the same application into ap-southeast-2 because many new users are in that region and are seeing high latency. A SysOps admin must ensure requests are automatically routed to the lowest-latency ALB for each user without changing the hostname. Which approach satisfies this requirement?

The correct approach is to use a latency routing policy. This policy routes user requests to the AWS Region that provides the lowest latency for the user, which directly addresses the requirement of automatically routing to the lowest-latency ALB without changing the hostname. You would create two latency records, one for each ALB in its respective region. Adding the new ALB as an additional value on the existing simple alias record would not provide latency-based routing; simple routing serves all records to all users. Switching to a geolocation routing policy would route users based on their geographic location, not necessarily their latency to a specific AWS Region. While it could be configured to approximate latency, it's not as precise as a dedicated latency policy. Multivalue routing returns multiple healthy records to the user, allowing the client to choose, but it does not guarantee the lowest latency and is typically used for client-side load balancing or failover.