Posted On: Jul 9, 2019

Amazon CloudWatch Anomaly Detection applies machine-learning algorithms to continuously analyze system and application metrics, determine a normal baseline, and surface anomalies with minimal user intervention. You can use Anomaly Detection to isolate and troubleshoot unexpected changes in your metric behavior.

You can apply CloudWatch Anomaly Detection on metrics in your account, including custom and AWS services metrics. CloudWatch Anomaly Detection will automatically determine a range of expected behavior, which you can optionally customize by specifying data exclusion periods, anomaly sensitivity, and daylight-savings time zone. You can create alarms to notify you when anomalies occur and visualize the expected behavior on a metric graph.

Get started by enabling Anomaly Detection via the AWS Management Console, AWS Command Line Interface, and AWS SDKs. Anomaly Detection is available in preview in all commercial AWS Regions except the Asia Pacific (Hong Kong) and China Regions. CloudWatch Anomaly Detection is priced per alarm. To learn more, please visit the CloudWatch Anomaly Detection documentation and pricing pages.