Anomaly detection service
The anomaly detection service employs an advanced neural network algorithm to identify data points of key metrics that are outside a metric's normal range of values. The service uses deep historical data and frequent model training to deliver accurate, precise, and near real-time results.
The following features use results from the anomaly detection service:
- The Anomalies tile automatically highlights anomalous entities.
- Smart View pages display anomalous metrics on the primary entity and on related entities.
- The action service supports anomaly triggers.
How it works
When the service finds enough historical data for a key metric, it starts training a model of the metric. When a new data point arrives, the service compares the data point with the model and then scores it with a value between 0 and 1. A score of 0.0 is 100% normal, and 1.0 is 100% anomalous. Perhaps unintuitively, almost all scores have to be greater than 0.99 to be considered anomalous.