Akka.NET: The Future of Distributed Programming in .NET
summary
Explains how Akka.NET maps the actor model into .NET with modules: akka core, akka.remote, akka.cluster (gossip membership), and akka.persistence (event sourcing to ~12 databases), and claims a receive actor can handle 6.7 million messages per second versus ~140 for Service Fabric in one benchmark. Actors guarantee single-threaded FIFO processing, default at-most-once delivery, location-transparent actor refs, TCP Helios transport, DeathWatch for failure detection, and cluster scaling via gossip.
AI-written summary · about
chapters (28)
- 0:00Intro
- 0:51High Expectations for Backend Developers
- 3:13Enterprise CRUD App
- 4:00Scenario 1: Surge in User Activity
- 4:40Obvious Solution: Sharding
- 6:04Brittle
- 7:39Scenario 2: Real-time User Interactivity
- 8:40Obvious Solution: Read-after- Write
- 10:37Why is Distributed Programming Hard?
- 14:54Distributed Systems 101
- 16:35Enter the Actor Model
- 17:36Origins
- 20:05Actor Model Core Concepts
- 25:32Akka.NET Modules
- 26:39What's an Actor?
- 31:15Actors live in a family tree
- 32:24Actors process one message at a time
- 35:01Actors share threads!
- 37:04Properties of Messages
- 41:03Messaging Patterns
- 42:33Messaging Protocols
- 43:56Communicate with Actors via Reference
- 44:45Actor References Have Transparent Locations
- 49:06Stateful Apps Serve Results from Memory
- 51:18State Makes Protocols Work
- 54:10CAP Theorem
- 54:50CAP Terminologies
- 55:37CAP & Protocols
related talks
39mAkka.net: The actor programming model
47mNetworks and Types — the Future of Akka
37mAkka Present and Future
44mDeath to Latency: Building Reactive, Cloud Native Apps with Akka NET
45mSo you want to be an actor? Lessons from implementing the actor model
50mActors can rule your DDD world
39mActors can rule your DDD world
41mHands on Introduction to Distributed Systems Concepts with Akka Clustering