Observability’s Sixth Sense: Detecting Anomalies in Metrics
summary
Shows a live workflow that wires OpenTelemetry into VictoriaMetrics queries and a backend that runs MAD, Z‑score and Prophet models, plus an MCP server that generates PromQL/MetricsQL queries and configs. Demonstrates detecting error‑span spikes and CPU anomalies, uses a 20% minimum deviation and consecutive‑anomaly “streaks” to cut alerts, and shows example scoring (50 ms baseline → 60 ms score ~15, 2,500 ms score ~95). Includes Grafana dashboards and a reproducible GitHub workshop with working demos.
AI-written summary · about
chapters (8)
- 0:00Introduction and speaker intro
- 1:50Why anomaly detection matters
- 7:48Key anomaly detection concepts
- 13:07Demo with traces and metrics
- 22:58MCP server for observability
- 27:37Workshop and resources
- 30:33Key takeaways and CNCF neurodiversity
- 34:06Q&A and closing
related talks
40mObservability’s Sixth Sense: Detecting Anomalies in Metrics
35mObservability’s Sixth Sense: Detecting Anomalies in Metrics
28mThe Sixth Sense: Finding “Dead” Metrics and Silent Alerts in Your Monitoring
38mWatch Everything, Watch Anything: Anomaly Detection
41mMetrics Applied
53mUnmask hidden issues with Observability superpowers!
50mHow To Improve PostgreSQL Monitoring and Observability With PMM
34mAnomaly Detection in Time Series from Scratch Using...