Question

A financial services company runs a customer-facing loan application API on AWS Lambda behind API Gateway. During business hours (9 AM to 6 PM on weekdays), users report intermittent latency spikes of 3-5 seconds on the first requests, which correlate with Lambda cold starts. Outside business hours, traffic is negligible. The team currently pays for provisioned concurrency 24/7 across all functions, which has significantly increased costs. The Solutions Architect must eliminate the cold-start latency during peak business hours while minimizing cost. Which approach best meets these requirements?

Accepted Answer

Configure Application Auto Scaling to schedule provisioned concurrency, scaling it up before business hours and down to zero on evenings and weekends.

Answer

Increase the Lambda function memory allocation to reduce initialization time and remove provisioned concurrency entirely.

Answer

Replace Lambda with Amazon ECS on Fargate tasks that run continuously to avoid cold starts.

Answer

Enable reserved concurrency on all functions and rely on API Gateway caching to mask cold-start latency.