Harnessing Chaos; the hidden ingredient behind building better systems through learning
summary
Production is hostile: hardware, latency, exhaustion and human error cause failure to be the norm, so the concrete claim is that chaos engineering builds measurable confidence in availability by running hypothesis-driven experiments against a defined steady state. Start small (limit the blast radius), automate experiments, run game days and stage-to-production iterations, and treat chaos as continuous disaster recovery. A Python-based open-source toolkit accepting JSON orchestrates experiments and captures actionable insights; banks already run these experiments and report useful, repeatable improvements.
AI-written summary · about
chapters (14)
- 0:00Introduction and context
- 8:33Personal background and road trip
- 12:42The reality of production
- 17:31Types of system failures
- 22:12Defining chaos engineering
- 26:49Complexity and systems theory
- 30:59Chaos engineering approach
- 34:35Running chaos experiments
- 37:17Blast radius and automation
- 42:10Rules of engagement
- 47:24Chaos in the enterprise
- 50:19Hard-earned advice
- 53:46Collaborative chaos
- 58:43Tooling and final takeaways
related talks
48mHarnessing Chaos; the hidden ingredient behind building better systems
37mChaos Engineering: The Fine Art of Breaking Stuff in Production on Purpose
27mChaos Engineering: The Path to Reliability
26mBreaking Systems for Resilience
21mThe Psychology of Chaos Engineering
29mThe Psychology of Chaos Engineering
18mChaos Engineering: Break It On Purpose
36mChaos Engineering in 2021