Azure Monitor AIOps Alerts with Dynamic Thresholds


Microsoft has released Azure Monitor AIOps Alerts with Dynamic Thresholds into public preview.  This new feature uses machine learning algorithms allowing administrators to setup monitors without manual thresholds.  Static alerts required the administraor to manually create seperate thresholds based on variations in the system and processes being monitored.  

Some Feature Highlights are:

Smart metric pattern recognition – Use ML Technology to identify patterns and a baseline for you, it will then be able to alert on any anomalies.

Scalable alerting – Create one alert rule that will be scalable across any system such as one CPU rule and ML will identify a baseline by machine or application.

Domain knowledge – Built-in ML algorithms to handle all this and find anomalies.

Intuitive configuration – users only required to select the sensitivity for deviations (low, medium, high) and boundaries (lower, higher, or both thresholds) based on the business impact of the alert in the UI or ARM API.