Building Evolutionary Architectures
summary
Claims that an incremental guided change approach is the core property of an evolutionary architecture and that cycle time equals evolutionary speed, backed by concrete tools and numbers: continuous delivery practices, JDepend unit tests to block cyclic dependencies, a CI performance pipeline that found a misconfigured SAN, and Netflix’s simian army chaos experiments that keep services resilient (Netflix stays up during an Amazon s3 outage). Examples include N 26’s 1.5 million customers and a CD pipeline that actually caught real regressions.
AI-written summary · about
chapters (16)
- 0:00Start
- 7:24GENERATIONS = CYCLE TIME
- 9:59BUILDING ARCHITECTURE
- 10:32WHAT IS ARCHITECTURE?
- 23:06FITNESS FUNCTIONS & CD
- 24:59TECHNICAL CHANGES
- 25:15DOMAIN CHANGES
- 25:54BIG BALL OF MUD
- 26:45LAYERED ARCHITECTURES
- 28:13MICROKERNEL
- 29:24MICROSERVICES
- 32:48PRINCIPLE-DRIVEN
- 33:38PRODUCTS over PROJECTS
- 35:01LAST RESPONSIBLE MOMENT
- 37:04Think like a town planner
- 37:48Technical Domain
related talks
42mBuilding Evolutionary Architectures: Principles and Practices
01h 03mEvolutionary Software Architectures
41mOrganising and Governing Evolutionary Architectures (Fitness Functions)
59mAgile Engineering: The Hard Parts
48mAgility and Architecture Together – Enabling a Fast Flow of Change
30mAn Evolutionary Approach to DevOps
43mEvolutionary Infrastructure Using Pipelines
01h 00mAgile Architecture