AIOps: Anomaly Detection with Prometheus and Istio

Explore advanced anomaly detection techniques for IT operations in this 37-minute conference talk by Marcel Hild from Red Hat. Learn how to enhance operational efficiency and intelligence in complex, agile IT environments using Prometheus and Istio. Discover methods for collecting and storing long-term metrics gathered by Prometheus, and analyze them on a large scale to extract trends, seasonality, and forecast expected values. Gain insights into integrating predicted metrics back into the Prometheus monitoring and alerting stack, enabling dynamic thresholding and anomaly detection. Follow along with a fully working demo that utilizes Istio's instrumentation and open-source tooling to implement these advanced AIOps concepts in your own Kubernetes clusters.