A company is migrating a data-processing application to AWS. Currently, users upload media files via a web portal; the web server saves uploads to NAS and sends a message to a processing server via a message queue. Each media file can take up to 1 hour to process. The number of files queued is much higher during business hours and decreases rapidly after hours. What is the MOST cost-effective migration design?

The correct option leverages Amazon SQS for reliable message queuing and an Auto Scaling group of EC2 instances for processing. SQS is a fully managed, cost-effective queuing service. Scaling the EC2 instances based on SQS queue length ensures that processing capacity dynamically adjusts to demand, minimizing costs during off-peak hours and handling peak loads efficiently. Storing processed files in Amazon S3 is cost-effective object storage. The first incorrect option uses Lambda for processing. While Lambda is serverless, the 1-hour processing time per file exceeds Lambda's maximum execution duration (15 minutes), making it unsuitable. The second and third incorrect options use Amazon MQ, which is more expensive and complex than SQS for this use case. They also propose Amazon EFS for storage, which is more expensive than S3 for processed file storage. The second incorrect option's approach of launching and shutting down EC2 instances for each message is inefficient and complex to manage compared to an Auto Scaling group.