Chaos + Observability + Resilience = Chaos Engineering

Abstract

The discipline of Site Reliability Engineering, more specifically, Chaos Engineering are becoming in a common practice for development teams. Netflix, Google, Gremlin and CapitalOne have done a great work promoting a premise: “Reliability” is the most important feature in software applications and “Chaos Engineering” is key for reaching this.

Observability and recently, Telemetry are key in all steps of Chaos Engineering. Validating hypothesis, analyzing steady state behavior, simulating real-world events and optimizing blast radius, which are critical in Chaos Engineering require observability in order to provide the expected value.

In this talk we are going to review how to observable data is critical in the Chaos Engineering underlines. We are going to explore the sources of observable data during the experiments and we are going to show observability in action in a Chaos Experiment using Gremlin and Google Cloud Platform.