Building an Anomaly Detector for High-Load from Python and Duct Tape
summary
Builds an anomaly detector for high-load services using Python and simple tools. Describes collecting many metrics (RPS, response time, open sockets, etc.), feature extraction and practical algorithmic approaches in Python to detect anomalies at scale for operational monitoring.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 2:56Anomaly detection problem
- 7:37Approach and development start
- 12:19System architecture
- 15:32False positive filters
- 21:33Data source optimization
- 25:04Automated metrics addition
- 30:09Team adoption and motivation
- 32:16Results and economic impact
- 33:21Future development
- 43:33Q&A
related talks
38mWatch Everything, Watch Anything: Anomaly Detection
45mAnomaly Detection in Time Series
26mAnomaly Detection System for the CERN Cloud
31mAnomaly Detection on OpenStack Logs using M.L.
19mThe Rise of AI in Anomaly Detection
19mAutomatic Metric Screening for Service Diagnosis
38mAnomaly detection in the cloud: how to detect issues in data [...]
01h 16mAnomaly Detection