Microservice Panel
summary
Conference panel argues containers and automated ops enable microservices but warn they move complexity into contracts, tracing and ops. Typical sizing anecdote is 30–40 services per 5–6 devs and one company reports 300 services with no message docs, which worked for internal teams but raises coupling risk. Recommended practices: API gateway, circuit breakers, caching only when necessary, avoid exposing databases, and apply Swagger + generative contract tests. One‑click or near‑one‑click deploy plus heavy production instrumentation proves crucial.
AI-written summary · about
chapters (10)
- 0:00Intro
- 2:12Introductions
- 5:10Size of Microservices
- 6:20Microservices and Data
- 10:58Polyglot
- 15:06Coupling
- 20:29Technology Advice
- 24:05Traceability
- 29:56Circuit Breakers
- 42:02Testing
related talks
43mPanel: Microservices, Where Did It All Go Wrong
01h 05mPanel: Were Microservices a Huge Mistake?
37mKnee Deep in Microservices
38mMicroservices, a bittersweet symphony
26mPros/Cons of Microservices
45mContinuously delivering containerized, highly available microservices
55mMicroservices, Where Did It All Go Wrong
26mCreating a scalable ecosystem of Microservices
