A startup is building a lightweight IoT backend that receives intermittent and unpredictable messages from devices in the field. The backend needs to process messages concurrently and scale precisely with incoming traffic to optimize costs. Management prefers a solution that minimizes operational overhead. Which of the following would provide the optimal balance of cost-efficiency and scalability?
Implement the backend with AWS Lambda to process messages with automatic scaling and no infrastructure management.
Use AWS Batch to manage message processing jobs, relying on its queuing mechanism to handle bursts in traffic.
Provision Amazon EC2 Spot Instances to handle the message load and reduce costs by bidding on spare computing capacity.
Deploy the backend using Amazon ECS on AWS Fargate to smoothly scale with workload changes, minimizing server management.