Alerting 101: Timeseries metric checks

This article discusses various types of alerts that can be used to monitor infrastructure components based on timeseries metrics and discrete events. Threshold alerts trigger when a metric goes above or below a user-defined threshold, with the option to include a time component to avoid false positives. Change alerts evaluate the delta or percentage change in a metric over a certain time interval, useful for identifying sudden changes in metrics with variable baselines. Outlier detection tracks deviations from expected group behavior and is suitable for isolating individual components that are failing to process work effectively. Anomaly alerts look for deviations from recent historical trends and can account for seasonality. Event alerts trigger on discrete occurrences, such as the completion of critical actions or unexpected activity. Composite alerts allow users to build more complex evaluation logic into their alert definitions by specifying that an alert fires if and only if a number of specific conditions are met.