Smarter Golden Signals - Using AIOps for Kubernetes Cluster Monitoring

Explore an innovative approach to Kubernetes cluster monitoring in this conference talk from KubeCon + CloudNativeCon North America 2022. Learn how Platform Engineers and SREs at Intuit tackled alert fatigue and improved incident detection using open-source solutions. Discover the implementation of numalogic, an AIOps anomaly detection engine, to analyze Prometheus metrics and derive baseline behaviors without requiring AI/ML expertise. Witness a live demonstration of the AIOps-based Prometheus metrics pipeline, showcasing real-time data collection, processing, and analysis. Gain insights into computing anomaly scores for individual components and aggregating them into a single cluster-wide score, ultimately reducing Mean Time to Detection (MTTD) during incidents and enhancing overall platform health monitoring.

Smarter Golden Signals! - Anusha Ragunathan & Venkata Gunapati, Intuit Inc