Panel: Architecture
summary
Panel argues specific tradeoffs: start with a monolith because it’s easier for small teams and split only when uneven growth forces independent scaling; splitting later often costs orders of magnitude more. Mentions concrete tools and designs: Kafka, zookeeper, site mesh 3, Kubernetes, JVM/Java, and cautions about inversion of control containers. Notes Netflix open source practice that drives tests and docs, and that the best climate move is not building the damn thing or at least ensuring can't get data out before committing to a vendor.
AI-written summary · about
chapters (8)
- 0:00Start
- 1:08James Lewis
- 15:45D3 Js
- 19:52Thoughts on Java Dying
- 21:00Java
- 28:19Kubernetes Etl
- 29:52What Is the Most Climate-Friendly Approach to Architecture
- 38:54Interdisciplinary Engineering
related talks
29mAsk the architects
32mPanel on Microservices
55mArchitectural Patterns for Rapid, Reliable, Frequent and Sustainable Development
50mEmbracing Microservices
43mMonolith? Microservices? Who cares?
01mHow npm split a monolith and lived to tell the tale
57mBuilding Systems that are #neverdone
33mFireside chat - Monoliths, Microservices & Modernization – Practical Strategies