Hypothesis Development Through Resilience Modeling
summary
Claims a concrete workflow: build a resilience model that maps business failure modes to scenarios and four control types, then turn scenarios into testable if‑then hypotheses. Uses a 2012 ELB config deletion as a cautionary example and a Kubernetes EKS storefront to show TLS certificate expiry, cache eviction under load, and missing session lookups. Notes customers report 65 incidents avoided when modeling first, and highlights gaps like missing testing controls.
AI-written summary · about
chapters (26)
- 0:00intro
- 0:22preamble
- 1:16phases of chaos engineering
- 2:32resilience capabilities
- 3:08amazon elb service event
- 4:56benefits of resilience modeling
- 5:44what is a system?
- 6:41a system in practice
- 8:00what is a system function?
- 8:23example system function
- 9:39what is a failure mode?
- 9:59example failure mode
- 11:12focus on what matters to your business...
- 11:51controlling failure
- 12:06types of controls
- 13:31example
- 14:28creating the hypotheses
- 15:20example
- 16:01let's put it in practice!
- 16:13customer application
- 17:23modeling "submit order"
- 19:21hypothesis for "submit order"
- 20:38phases of chaos engineering
- 21:18key takeaways
- 22:38new! resilience space at community.aws
- 22:54thank you!
related talks
55mModels For Resilience
43mResilience Engineering Mythbusting
40mResilience Testing - Failure is an Option
30mResilience Engineering Mythbusting
25mBuilding a highly scalable system to assure Business Resilience
11mResilience engineering mythbusting
16mEngineering Failure-Resilient Systems
22mResilience Is Not Enough