A company has deployed a model to an Amazon SageMaker endpoint and needs to record and monitor every API invocation for that endpoint, plus receive a notification when the number of invocation events exceeds a threshold. Which solution satisfies these requirements?

The correct solution uses AWS CloudTrail to log all API calls, including SageMaker endpoint invocations. CloudTrail provides a comprehensive record of actions taken by a user, role, or AWS service, making it ideal for auditing and monitoring API activity. This log data can then be sent to Amazon CloudWatch Logs, allowing you to create a CloudWatch dashboard for visualization and set up CloudWatch alarms to trigger notifications when the number of invocation events exceeds a specified threshold. SageMaker Debugger is designed for debugging and profiling training jobs, not for monitoring endpoint invocation API calls. While it can track inference data, it's not the primary tool for logging API calls or setting up invocation count alerts. Simply adding the Invocations metric to a CloudWatch dashboard and creating an alarm will monitor the number of invocations, but it won't provide the detailed logs of each individual API call, which CloudTrail does. The requirement is to "record and monitor every API invocation," which implies detailed logging.