Reactive all the Things
summary
A fast realtime dashboard at Netflix hits severe GC and latency from heavy map/filter/reduce chains that allocate many intermediate arrays, so RxJS observables replace array pipelines with stream processing and eliminate extra allocations; a socket observable is built that opens on subscribe, emits messages, sends subscribe/unsubscribe over a multiplexed WebSocket, uses retry and published ref counted semantics to reconnect and close automatically, and Angular 1 helpers like $createObservableFunction and observable.times enable mousemove-driven staggered animations while requiring explicit disposal to avoid leaks.
AI-written summary · about
chapters (11)
- 0:00Introduction and speakers
- 1:22Background and careers
- 3:46Why functional programming
- 6:27Netflix performance issues
- 9:16Introduction to RxJS
- 11:23Solving complex UI problems
- 13:56Reactive programming concepts
- 16:12Angular 1 examples
- 17:37Observable operators
- 22:12Mouse move animation demo
- 26:44Conclusion and Q&A
related talks
27mReactive Programming in Node
43mReactive JavaScript at Netflix, Microsoft and the World
36mRxJS: A Better Way to Write Frontend Applications
24mAdding Even More Fun to Functional Programming With RXJS
37mStreaming and event-based programming using FRP and RxJS
45mReactive Frontends with RxJS and Angular
28mWatch out: observers are here to stay
30mAsynchronous JavaScript
