The Guidance sends events to an EventBridge event bus and provides a rule that matches those events. You can also add an Amazon SNS topic as a target to the rule. These aspects provide visibility into actions and let you integrate the actions with downstream systems to produce notifications or enable additional automation. Additionally, this Guidance supports decoupled integration with other systems. Finally, CloudWatch monitors the infrastructure and provides logs and metrics that aid troubleshooting of potential issues.

AWS Identity and Access Management (IAM) lets you set up granular permission policies and short-lived access through roles (rather than hard-coded credentials). By using IAM policies to implement the principle of least privilege for the Lambda function used in this Guidance, you can limit unauthorized access.

This Guidance uses an Amazon SQS queue between event sources and the Lambda function that consumes those events. This enables Lambda to perform retries without event loss in the case of failure. Additionally, EventBridge is a fully managed and internally resilient service that helps you avoid any single point of failure.

This Guidance uses event-driven services like EventBridge, Amazon SQS, and Lambda. This supports asynchronous, responsive behavior so that the solution can react and implement actions promptly. For example, Amazon EC2 instances stop (or terminate) within low-single-digit seconds of their scheduled expiration. This is more effective than a polling-based design, which would check instances for expiration at set intervals, resulting in imprecise implementation.

EventBridge, Amazon SQS, and Lambda provide a pay-as-you-go pricing model, meaning you pay only for resources used. Because this solution remains idle for extended periods, these cost-elastic services offer significant savings by eliminating compute costs during times of inactivity.

This Guidance uses services that support event-driven, asynchronous, and intermittent behaviors. For example, Lambda only uses compute resources when necessary. And as a shared service, Amazon SQS does not dedicate compute to this Guidance specifically. These aspects make this solution resource efficient, supporting sustainability.