Resilience As A Continuous Delivery Enabler
summary
Argues that optimizing for MTBF (robustness) traps organisations in long lead times and “risk management theatre,” citing government teams spending 4–6 weeks preparing end‑to‑end tests whose test doubles ran 700× slower at 1% production load. Proposes instead to optimize for MTTR via operability: telemetry, version‑controlled config, feature toggles and an adaptive architecture. Gives examples: a digital media team moved to daily deploys within a month by exposing pipeline metadata to change governance; MTTR falls from days to minutes/hours, cutting failure cost and enabling Continuous Delivery.
AI-written summary · about
chapters (26)
- 0:00Introduction
- 3:18Mean vs MTTR
- 5:16Endtoend testing
- 6:46Cabs
- 8:00Change freezes
- 9:08Why deployments dont prevent outages
- 9:57A real story
- 12:07Risk management theatre
- 13:05Optimize for robustness
- 14:35Stop asking questions
- 16:16People craving certainty
- 17:06The world is messy
- 18:08Dual value streams
- 21:29Optimize for resilience
- 22:39Graceful extensibility
- 24:23Operability
- 27:30Improve awareness
- 29:32Adaptive architecture
- 31:05Instant response
- 32:48Risk management
- 34:21Testing
- 36:28Change Management
- 39:19Feature Toggles
- 41:14Summary
- 43:06Automation vs Manual
- 44:41Operability vs Functional Work
related talks
40mSustainable Continuous Delivery: Patterns for rapid and reliable Releases
22mResilience Is Not Enough
25mBuilding a highly scalable system to assure Business Resilience
34mDevOps: Next
34mToward continuous resilience
19mThink Prod Up Front: Shifting Left on Production Excellence
36mWhat do we know about continuous delivery – and what would we like to know
47mWhere the Security Rubber Meets the DevOps Road