Go distributed (and scale out) with Actors and Akka Clustering
summary
Explains how William Hill modernises a betting pipeline using Akka Cluster inside a data platform called Omnia, including a custom clustering strategy, node membership via the gossip protocol, and how cluster leadership, vector clocks and failure detectors converge. Gives concrete load targets—5 million price updates per minute and 300,000 active sessions—and shows why you must design clustering, sharding and supervision for availability, elastic scale-out and isolated failure handling.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:39Agenda
- 2:16Requirements
- 3:49Reactive Manifesto
- 6:11Distributed System
- 8:37Actors
- 14:22Vertical scalability
- 15:36Reactive programming
- 17:12Origin to scalability
- 17:39How Akka is organized
- 18:02Akka Remote
- 18:45Akka Cluster
- 19:52Fault Tolerance
- 21:38Gossip Exchange
- 22:21Failure Detection
- 23:17Configuration
related talks
59mActors in clusters: why, when and how
59mLife of Actors in a Cluster: Why, When and How
41mHands on Introduction to Distributed Systems Concepts with Akka Clustering
01h 14mAkka.NET: The Future of Distributed Programming in .NET
01h 59mAkkaCluster and AkkaTyped DeepDive Into the ActorModelWith a BitofEyeCandy.
45mIs there anybody out there?
27mFour Principles of Reactive Architectures
53mDistributed Eventsourcing with Akka