From Observability to Performance in Kubernetes - Metrics, Analysis, and Optimization

Explore a comprehensive conference talk that delves into transforming Kubernetes cluster metrics into tangible user experience improvements. Learn to identify critical performance parameters, implement effective metrics collection using tools like Prometheus, Grafana, and kube-burner, and gather data from running clusters without operational disruption. Master the process of translating collected metrics into actionable insights, detect performance bottlenecks, and implement targeted optimizations for enhanced service performance. Designed for Kubernetes networking developers, administrators, SREs, and DevOps engineers, gain practical knowledge about improving distributed system performance through systematic observation and analysis.