Beyond Fault Tolerance
summary
Explains how the Actor model with Akka on the JVM implements reactive-system traits: message-driven isolation, supervision and supervisor restarts child semantics that replace messy in-place recovery. Demonstrates a ping‑pong actor pair and a coffee‑machine where an out‑of‑coffee exception triggers a restart and the user still receives coffee, proving resilience over fault tolerance. Shows Rx-style backpressure handling for streams, claims Scala can host millions of actors, and gives concrete code links and restart strategies that actually work in the demo.
AI-written summary · about
chapters (16)
- 0:00Start
- 1:49Reactive Amsterdam meetup
- 2:47Outline
- 3:17Reactive Programming & its benefits
- 4:16The World is a Streaming Place
- 5:08One challenge: different speed of producer and consumer
- 6:57Reactive Programming example Relava on Android
- 11:22The Reactive Principles
- 12:03The Reactive Manifesto, 2013
- 13:36Reactive System traits (the Reactive Manifesto)
- 14:10Reactive Programming vs Reactive Systems
- 14:58From Principles to implementation
- 16:00Why Scala?
- 18:33Simple Component Pattern
- 20:58The benefits of Asynchronous Messaging
- 26:41Let it Crash Pattern
related talks
50mNext Generation Actors With Akka
27mFour Principles of Reactive Architectures
18mReliable Actors with Retry Orchestration
33mA Reliability Benchmark for Actor-Based Server Languages
42mTesting Akka actors in smart way
17mYou asked for it, you got it: type safe Actors
22mAdvanced Akka features
32mAdopting Actors... an epic tale of loss and learning