Improving Business Resiliency with Chaos Engineering
summary
Prime Video adds chaos experiments into release pipelines, running automated scaling experiments three times a week in production and combining load, stress and failure injection to validate launches like Bundesliga and Grand Tour (8M viewers). One experiment injects 2s latency and injected 2s latency found 6s timeout bug after a 1,000‑request load, preventing a likely outage. Playbooks set specific goals (100% availability, detection and blast‑radius time), distribute ownership, and measure KPIs to drive culture change.
AI-written summary · about
chapters (6)
- 0:00Intro
- 1:33Let's start with questions
- 9:13Where to start
- 25:45How to make meaningful progress?
- 43:26Journey takeaways
- 44:32Looking ahead
related talks
26mBreaking Systems for Resilience
38mChaos Engineering: A Step Towards Resilience
31mPlanning for Failure using Chaos Engineering
18mChaos Engineering: Break It On Purpose
23mHow to tame chaos effectively?
29mImprove Resilience with automated chaos engineering
35mImprove resilience with automated chaos engineering
48mChaos Engineering: The Path to Reliability