AWS Certified Solutions Architect Professional SAP-C02 Practice Question

The correct answer is to use AWS Step Functions to orchestrate a series of AWS Lambda functions. This approach directly addresses the primary challenge, which is the potential for the process to exceed the 15-minute maximum execution time of a single AWS Lambda function. By breaking the long-running process into smaller, sequential tasks, each can be executed by a separate Lambda function within its time limit. AWS Step Functions manages the state, sequencing, and error handling between these functions, effectively creating a long-running, resilient, and serverless workflow. This architecture is event-driven, triggered by the S3 event, and is highly cost-effective because you only pay for the Step Functions state transitions and the actual compute time of the Lambda functions, which aligns perfectly with the unpredictable, bursty workload.

Using a single Lambda function is incorrect because it cannot handle processes that run longer than its 15-minute maximum timeout. While a Dead Letter Queue (DLQ) can handle failures, it doesn't solve the fundamental execution time limitation.

Using AWS Fargate is a plausible but less optimal solution. While Fargate can handle long-running containerized tasks without a 15-minute limit, it is generally less cost-effective for highly sporadic and unpredictable workloads compared to the pay-per-execution model of Lambda and Step Functions. The Lambda and Step Functions combination is more aligned with a purely event-driven, function-based modernization approach.

Using an SQS queue with an EC2 Auto Scaling group is not a serverless approach. This pattern requires managing EC2 instances, which is what the company is trying to modernize away from. It is less efficient for handling sudden bursts due to the time it takes for new EC2 instances to launch and is not as cost-effective during long idle periods.