A legacy application migrated to EC2 behind an ALB and API Gateway needs deployments that minimize user disruption and allow rapid rollback with minimal application changes. Which approach meets these goals?

The correct answer is to deploy the new version as a separate parallel environment and use API Gateway canary deployments. This approach allows for a controlled rollout of the new application version by gradually shifting a small percentage of traffic to the new environment. If issues arise, traffic can be immediately shifted back to the old environment, minimizing user disruption and enabling rapid rollback without application changes. Deploying a separate parallel environment and switching DNS alias records (blue/green deployment) is a valid strategy but typically involves a full cutover, which might cause a larger impact if issues are discovered after the switch. Adding a new target group and configuring API Gateway to route traffic in phases is similar to canary but might require more complex configuration within the ALB and API Gateway to manage the phased rollout effectively. Adding a new target group and immediately forwarding all traffic to it is a direct cutover, which carries the highest risk of disruption if the new version has problems.