Posted On: Oct 17, 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, reducing the mean time to detect and resolve operational issues.

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

It is easy to get started with Anomaly Detection. In the CloudWatch console, go to Alarms in the navigation pane to create an alarm based on Anomaly Detection, or start with Metrics to overlay the metric’s expected values onto the graph as a band. You can also enable Anomaly Detection using the AWS Command Line Interface, AWS SDKs, or AWS CloudFormation templates. Anomaly Detection is now generally available in all commercial AWS regions, including AWS China (Beijing) region, operated by Sinnet, and AWS China (Ningxia) region, operated by NWCD. To learn more, please visit the CloudWatch Anomaly Detection documentation and pricing pages.