Building Reactive Systems with Reactor
summary
Explains how Spring Reactor (an implementation of Java 9 reactive streams) and reactor-netty replace thread-blocking HTTP calls with nonblocking Mono/Flux pipelines, showing RestEasy’s MonoRxInvoker wrapped by a trace invoker in PayPal’s raptor framework, how subscribeOn/publishOn and scheduler pools control threads, how backpressure requests (s.request(n)) look in code, and reports noticeably better CPU and memory efficiency plus a strategy to limit DB overload.
AI-written summary · about
chapters (18)
- 0:00Introduction and agenda
- 1:02Why reactive systems
- 3:28What is reactive programming
- 4:11History of reactive systems
- 5:36Why Reactor over alternatives
- 6:19Reactor core library
- 8:26Building blocks of Reactor
- 9:08Publishers Flux and Mono
- 10:36Operators in Reactor
- 12:00Schedulers and concurrency
- 13:22Focus for developers
- 14:04Code examples Flux and Mono
- 15:08Backpressure in Reactor
- 16:10Schedulers in code
- 17:57Non-blocking HTTP calls
- 19:22Learnings from building systems
- 21:32Advanced Reactor features
- 22:35Q&A session
related talks
47mReactive Spring
01h 21mReactive Web Applications
01h 32mBuilding Reactive applications with Reactor and the Reactive Streams standard
51mReactor 3, the reactive foundation for Java 8 (and Spring 5)
49mReactive Spring
41mThe definite guide to reactive programming in Java
53mRefactor to Reactive with Spring 5 and Project Reactor
51mReactive Spring