Just-in-time Architecture
summary
A games team moves from a monolith to microservices, discovers an accidental distributed monolith with availability coupling (three 99% services yield ~97% system availability — about 11 days/year downtime), then adopts an event-driven design using TypeScript code, event sourcing and a transactional outbox to guarantee events, and an orchestration-style process manager to issue commands, handle retries and compensating transactions while accepting eventual consistency and added observability.
AI-written summary · about
chapters (9)
- 0:00Just-in-time Architecture intro
- 1:29Games company’s journey
- 8:05Monolith to microservices pitfalls
- 14:24Why distributed monoliths fail
- 15:26Event-driven architectures
- 22:50Event-driven communication
- 31:12Scaling architecture evolution
- 33:38Checkout workflow live code
- 35:31Q&A
related talks
49mJust in Time Architecture
55mArchitectural Patterns for Rapid, Reliable, Frequent and Sustainable Development
01h 03mMicroservices vs The Distributed Monolith
40mCAPtivating architecture: Navigating Distributed Systems and Microservices
48mIT Architecture : All your micro-services are wrong PERIOD!
27mThe technical journey to microservices
59mHow to Escape The Distributed Monolith
01h 08mEvent Driven Architecture – Enabling Microservices
