Time Series Anomaly Detection (TSAD) is the process of detecting abnormal behavior — anomalies — in time series data. Time series data is data that captures the value of a metric at a point of time — for example, number of errors in a TV stream, CPU utilization of a server, temperature recorded by a data center sensor, and more.

Devo solves the machine-scale TSAD problem by applying machine-learning techniques to learn unique profiles of what is “normal” per metric being measured.