Architectural Patterns for Rapid, Reliable, Frequent and Sustainable Development
summary
Argues that architectures must enable deploy multiple times a day with lead time under 15 minutes, comparing traditional monolith, modular monolith and microservices by concrete tradeoffs. Shows monoliths growing to multi-gigabyte WARs, long test runs and slow CI; modular monoliths use vertical slices and a customer-api module to avoid rebuilds; microservices require sagas, CQRS or API composition for cross-service operations, consumer-driven contract tests, protobuf/gRPC for APIs and canary deployments to validate releases. Choices trade runtime coupling, eventual consistency and team autonomy.
AI-written summary · about
chapters (10)
- 0:00Introduction and speaker background
- 1:55Why rapid delivery matters
- 8:15Development-time architectural requirements
- 16:45Monolithic architecture explained
- 21:19Challenges of large monoliths
- 26:14Modular monolith variations
- 34:21Microservice architecture overview
- 41:01Complexities of distributed systems
- 44:36Key takeaways and architecture evolution
- 45:40Q&A session
related talks
01h 01mArchitecting for rapid, frequent, reliable delivery of software.
39mMinimal Viable Architecture
43mMonolith? Microservices? Who cares?
43mModular Monoliths: The Architecture That Scales w Your Domain
01h 03mMicroservices: Decomposing Applications for Testability and Deployability
37mEvolutionary Architecture & Microservices
01h 00mAgile Architecture
47mMicroservices? Monoliths? An Annoying Discussion!