Big microservices, small team a painful story
summary
Company migrates a monolith into dozens of business-capability services over seven years, then loses 80% of developers after a buyout and must run everything with one team. The team keeps services independently deployable with CI, Octopus Deploy, Cake, Sonar and library-scans, tracks ~160 services and platform state in a spreadsheet, and ports libs to .NET Core/.NET 6. Key pain: leftover legacy code and diverse pipelines — lesson: remove old code and standardize deploys.
AI-written summary · about
chapters (13)
- 0:00Introduction and speaker intro
- 1:08Journey from guerrilla SOA to microservices
- 8:24Conway's law and business capabilities
- 12:41Company sold and team collapse
- 15:52Reverse Conway's law and team restructuring
- 19:05Surviving with a single team
- 23:54Focus on single business capability
- 27:45Diverse codebases and technical debt
- 34:27Tracking system state and documentation
- 41:38Current state and future steps
- 42:19Q&A: Defining business capabilities
- 45:04Q&A: New vs. fixing old code
- 47:56Q&A: Codebase diversity trade-offs
related talks
35mSeven years with microservices: a critical review
01h 01mMicroservices Gone Wrong
48mMicroMonolith
29mMoving from A and B to ~150 microservices, the journey and learnings
48mThirty months of microservices
53mSolving technical debt of a microservices architecture
54mFrom a distributed monolith to a microservices solution
23mOn the road from Monolith to Micro
