Models For Resilience
summary
Explains resilience as measurable system performance and maps three practical models: Woods’s four capabilities (robustness, rebound, graceful extensibility, sustained adaptability), Dickerson’s service hierarchy with four golden signals and incident/postmortem practices, and a hexagonal socio‑technical model tying software, infra, process, people, culture and goals. Cites Google’s cheap‑hardware + fast‑replace racks, shows failure math (1-in-100 per component → near‑certainty at scale), warns against blame cultures (Telstra case), and recommends OpenTelemetry, Kubernetes and desired‑state/GitOps patterns.
AI-written summary · about
chapters (22)
- 0:00Resilience in distributed systems
- 1:55Defining resilience
- 5:35Safety management vs resilience
- 12:47Models for resilience engineering
- 13:34Woods's four concepts
- 17:21Robustness handling known events
- 18:05Rebound after trauma
- 18:32Graceful extensibility
- 19:43Sustained adaptability
- 21:38Service reliability hierarchy
- 23:05Observability and golden signals
- 24:17Incident response and post-mortems
- 26:53Testing and capacity planning
- 28:42Development for resilience
- 30:59Hexagonal model for systems
- 36:19Sociotechnical system components
- 40:08Case study: system failure analysis
- 45:34Summary of resilience models
- 47:15Q&A: complexity and business alignment
- 48:28Q&A: tools for resilience
- 50:22Q&A: trade-offs and alignment
- 54:57Closing remarks
related talks
44mFrom here to resilience - a travel guide
40mThe Resilience Business Case: Three Case Studies
22mResilience Is Not Enough
16mResilient by Design: Architectures for Crisis Response & Recovery
43mResilience Engineering Mythbusting
30mResilience Engineering Mythbusting
58mResilient by Design
23mHypothesis Development Through Resilience Modeling