Question

A company runs a customer-facing API backed by an AWS Lambda function that uses provisioned concurrency to avoid cold starts. During unpredictable traffic spikes, users experience latency because requests exceed the configured provisioned concurrency and spill over to on-demand invocations. A SysOps administrator must automatically increase provisioned concurrency when utilization is high, without over-provisioning during quiet periods. Which approach meets these requirements with the LEAST custom code?

Accepted Answer

Register the Lambda function's alias as a scalable target with Application Auto Scaling and configure a target tracking scaling policy on the ProvisionedConcurrencyUtilization metric.

Answer

Create a CloudWatch alarm on the Duration metric and configure the alarm action to invoke a Lambda function that calls PutProvisionedConcurrencyConfig.

Answer

Create an EventBridge scheduled rule that runs every five minutes to increase provisioned concurrency during business hours.

Answer

Enable CloudWatch anomaly detection on the ConcurrentExecutions metric and attach an SNS action to notify the operations team to adjust concurrency manually.