Pub crawling in Orleans: exploring the Actor model
summary
Demo builds a global pub‑crawl backend with Orleans 4 with .NET 7 using the virtual actor model to represent crawlers, venues, beer scores and leaderboards; state persists via Orleans persistence (in-memory for demo, ADO.NET / Azure / DynamoDB supported). The demo runs a Blazor front end, SignalR hubs, and a simulator that ramps to 1000 simulated crawlers, shows timers/reminders, smart caching, and graceful silo failover — the system works and rebalances actors across silos.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 0:27Pub Crawl scenario
- 5:01Scaling distributed systems
- 6:25Orleans and virtual actors
- 7:29Actor model basics
- 11:23Orleans vs Akka.NET
- 16:20Crawler grain implementation
- 22:21Orleans backend setup
- 27:15Cluster testing
- 31:33Pub Crawl web app
- 34:02Rating beers with grains
- 39:23Timers in grains
- 42:32Leaderboard caching
- 44:42Venue attendance tracking
- 46:50SignalR integration
- 50:00Load simulation
- 51:26Cluster monitoring
- 52:33Real-world scalability
- 54:41Q&A
related talks
01h 01mHow to build real-world applications with Orleans
40mActors in .NET, 2025 edition
59mIntroduction to project Orleans for your distributed applications
52mModern Observability and Event Driven Architectures
01h 05mModern Observability and Event Driven Architectures
53mOrleans under the hood
46mState or events? Which shall I keep?
01h 03mDapr for building distributed .NET Core applications