Tutorial: Full-Stack Observability on a Budget: A Guide to Strategic Sampling and Da... Pavol Loffay

This tutorial covers strategies for achieving full-stack observability while keeping costs in check, presented by Pavol Loffay from Red Hat. The session explores a holistic framework for managing observability expenses, discussing head-based, probabilistic, and tail-based sampling along with their respective trade-offs in cost, computational overhead, and data fidelity. The speaker also addresses the often-overlooked hidden costs associated with tail sampling and provides insights on identifying and eliminating waste in telemetry data. Attendees will gain practical knowledge on designing a cost-effective observability strategy in Kubernetes using OpenTelemetry, selecting appropriate sampling methods, and obtaining better visibility into observability spending.