So you want to be an actor? Lessons from implementing the actor model
summary
System runs in production on Akka.NET with F# and C# and proves the actor model scales and simplifies architecture: files become one actor each, removed file dispatchers, and routing complexity drops. Persistence moves to the aggregate-root to avoid querying 20–30 child actors, accepting eventual consistency (rare duplicate uploads). Avoid ask (huge latency); use aggregation/state transitions and stashing instead. Use Akka persistence, snapshots and reminders (or Quartz) for long‑running workflows, and prefer symmetrical clustered routers over fragile singletons and automatic split‑brain killing.
AI-written summary · about
chapters (11)
- 0:00Intro
- 7:35ObjectOriented
- 13:32File Dispatcher
- 15:42Persistence
- 18:52Active state
- 20:55Mediaset monitor
- 23:08Periodic reminders
- 29:08Using actor model in a cluster
- 31:23Applying domain driven design
- 34:02Clusters
- 35:53Splitbrains
related talks
44m10 Lessons From Implementing The Actor Model
48mReliable messaging in the world of actors
52mBuilding Concurrent, Fault Tolerant, Scalable Applications in F# using Akka Net
01h 14mAkka.NET: The Future of Distributed Programming in .NET
59mLife of Actors in a Cluster: Why, When and How
29mTHINGS I WISH I KNEW WHEN I STARTED WITH AKKA
51mBuilding concurrent systems using Akka.NET
50mNext Generation Actors With Akka