From Reactive Extensions to Reactive Streams
summary
A clear lineage from Rx.NET and async enumerables to the Reactive Streams standard, with concrete trade-offs: lack of back-pressure in Rx versus back-pressure via request(n) in Reactive Streams; allocator costs like ~88 bytes per Task (up to ~300 bytes on thread-pool) leading to millions of allocations in hot loops; Reactive Streams TCK is ~100 compliance checks. Demonstrations use Akka.NET Streams: a fused, materialized pipeline, a TCP echo server/client, SQL-backed live event sourcing, and SourceRef spanning streams across processes — all implemented and running.
AI-written summary · about
chapters (39)
- 0:00Introduction
- 0:38Controversial ideas
- 1:05Outline
- 2:01API vs SP
- 2:30Tradeoffs
- 4:02API based libraries
- 5:11SPI
- 5:33Design Decisions
- 8:06Reactive Extensions
- 11:16Async enumerable
- 11:32Async continuation
- 12:28Publishsubscribe model
- 12:58Limitations
- 13:16NARS
- 14:15Tasks
- 16:55Slow continuation
- 17:31Batching
- 18:54Sideline
- 19:58Dynamic PushPull
- 22:02Demand
- 23:42The Magic
- 24:06Syncing
- 24:43Task
- 25:47AirSocket
- 26:29Test Compatibility Kit
- 27:21Service Provider Interface
- 28:09Server Side Extensions
- 31:14Source Flow Sync
- 31:54Demo
- 34:37Materialization Process
- 36:18Interpreter Pattern
- 37:44Materialization
- 41:09Reactive Stream IO
- 49:10byte strings
- 50:42beautiful flow
- 52:45graph DSL example
- 55:15Persistence
- 56:10Simple Actor
- 58:05Stream Rep
related talks
56mThirst-quenching Streams for the Reactive Mind
46mAkka Streams & Reactive Streams in action
01h 01mThirst Quenching Streams for the Reactive Mind
49mThirst Quenching Streams for the Reactive Mind
01h 32mBuilding Reactive applications with Reactor and the Reactive Streams standard
01h 28mIntroduction to Reactive Programming
48mReactive Streams - Beware of Dangers!
01h 21mReactive Web Applications