Which solution will meet these requirements with the LEAST effort?
Deploy multiple instances for each endpoint in a VPC that spans at least two Regions.
Use the SageMaker auto scaling feature for the hosted recommendation models.
Deploy multiple instances for each production endpoint in a VPC that spans least two subnets that are in a second Availability Zone.
Frequently generate backups of the production recommendation model. Deploy the backups in a second Region.
Explanations:
Deploying multiple instances across two Regions increases complexity and requires significant effort in managing cross-region networking and latency. This option may also lead to higher costs without necessarily achieving the desired RTO of 5 minutes.
While the SageMaker auto scaling feature can help manage load, it does not address high availability directly. Auto scaling adjusts capacity based on demand but does not ensure uptime in the event of a failure or availability zone outage, which is critical for achieving the specified RTO.
Deploying multiple instances across subnets in a second Availability Zone provides redundancy within the same region. This setup allows for automatic failover in case of an Availability Zone outage, ensuring that the system can recover quickly and meet the RTO of 5 minutes with minimal effort.
Frequently generating backups and deploying them in a second Region is a valid disaster recovery strategy but does not ensure high availability. The RTO of 5 minutes is unlikely to be met, as restoring from backups typically takes longer than a few minutes, making this approach unsuitable for the requirement.
If memory serves me right, the answer is:
Deploy multiple instances for each production endpoint in a VPC that spans least two subnets that are in a second Availability Zone.