Morgoth
A flexible anomaly detection framework for Kapacitor using fingerprinting algorithms and lossy counting.
What is Morgoth?
Morgoth is an anomaly detection framework designed to integrate with Kapacitor for identifying unusual patterns in streaming metrics data. It works by maintaining a dictionary of normal data windows using fingerprinting algorithms and the Lossy Counting Algorithm, flagging windows that deviate significantly from learned norms. The framework supports multiple detection methods and consensus-based voting to improve accuracy.
Developers and data engineers working with InfluxData's TICK stack (particularly Kapacitor) who need customizable, real-time anomaly detection for operational metrics or monitoring pipelines.
Morgoth offers a modular, algorithm-agnostic approach to anomaly detection that can be tailored with different fingerprinters, and its efficient memory management via LCA makes it suitable for long-running streaming applications without unbounded memory growth.
Overview
Metric anomaly detection
Use Cases
Best For
- Detecting anomalies in real-time CPU, memory, or application metrics from Telegraf
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.