How to tame chaos effectively?
summary
Explains practical chaos engineering: run controlled failures starting in dev, instrument steady‑state metrics and automate experiments as a gated step. Recommends attack templates (network +100 ms, CPU, memory), an example load of 20,000 QPS, and over 50 tools like Chaos Monkey and Grammine; shows Netflix moved to cloud after a three‑day outage and recommends running experiments at least once per quarter. Gives concrete stop conditions, backlog actions, and business impact numbers (Apple $25M/12h, Delta $150M/5h).
AI-written summary · about
chapters (9)
- 0:00Introduction and Speaker Background
- 0:46Understanding Chaos Theory
- 2:47Chaos Engineering in Practice
- 4:03History and Evolution of Chaos Engineering
- 5:49Principles and Implementation of Chaos Engineering
- 10:26Tools for Chaos Engineering
- 15:38Common Traps and Tips for Chaos Engineering
- 18:48Convincing Leadership and Benefits of Chaos Engineering
- 22:04Conclusion and Final Thoughts
related talks
18mChaos Engineering: Break It On Purpose
46mDesigning a System for Chaos
48mChaos Engineering: The Path to Reliability
37mChaos Engineering: The Fine Art of Breaking Stuff in Production on Purpose
38mChaos Engineering
36mChaos Engineering
52mGetting started with Chaos Engineering
38mChaos Engineering: A Step Towards Resilience