Managing to Your SLO Amidst Chaos

Explore strategies for maintaining Service Level Objectives (SLOs) in chaotic environments through this 40-minute conference talk from YOW! 2022. Delve into Honeycomb's approach to handling incidents, implementing chaos engineering, and fostering a reliability-focused engineering feedback loop. Learn how to measure reliability, stay within SLOs, validate expectations, and conduct experiments in production. Discover techniques for balancing speed and reliability, and gain insights from real-world examples of both successful and unsuccessful experiments. Benefit from practical advice on quantified reliability, incident management, and architectural design for improved service performance.

Intro
Our confidence recipe
Measuring reliability
How to stay within SLO
Validating our expectations
Experimenting in prod
Not every experiment succeeds
Fast & reliable: Pick both!
Outro
Q&A