DevOps and Microservices Better Together
summary
Argues that achieving rapid CI/CD requires a modular, well‑encapsulated, loosely‑coupled architecture and shows concrete moves: replace synchronous RPC with queue‑based messaging to avoid cascading failures, adopt separate read/write APIs (CQRS), own data per bounded context with no cross‑database referential integrity, and split code into per‑service repositories and pipelines. Cites use of Git, TeamCity, unit tests and an event‑driven pattern, and reports a roughly 100% throughput improvement after breaking a monolith into service repos and pipelines.
AI-written summary · about
chapters (8)
- 0:00DevOps and microservices intro
- 5:51From chaos to Agile DevOps
- 9:45Monoliths vs microservices
- 14:40Coupling and runtime deps
- 18:32Data ownership and contexts
- 21:22Vertical slicing and CQRS
- 25:35Case study scaling microservices
- 28:22Future of DevOps microservices
related talks
45mContinuously delivering containerized, highly available microservices
57mForking the Future: Embracing Microservices Evolution
50mEmbracing Microservices
37mKnee Deep in Microservices
01h 03mMicroservices: Decomposing Applications for Testability and Deployability
23mDevOps and Legacy Code
27mMicroservices Are ‘Easy’, Dependencies Are Hard
55mArchitectural Patterns for Rapid, Reliable, Frequent and Sustainable Development