Building high available systems with Akka Distributed Data
summary
Overview of building highly available systems using Akka Distributed Data: explains how state-based CRDTs (including ORSet and PN-counters) gossip full state or Delta CRDTs between nodes, with a Replicator actor API for Get/Update/Delete and configurable read/write consistency (local, quorum, all). Demo runs two nodes (ports 880/881) with HTTP writes that converge via gossip; updates apply locally then propagate, can timeout without rollback. Notes store all data in memory (optional LevelDB persistence) and suit eventual‑consistency use cases like service discovery or collaborative lists, not strict consistency or huge datasets.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 0:39Overview
- 1:15What does it mean
- 1:51Example of a distributed system
- 2:52Cap Theorem
- 4:04Consistency
- 6:59CoDT
- 9:22Vector Clock
- 14:47Akka Distributed Data
- 16:32Replicator
- 17:26Interface
- 22:05Consistentity
- 28:32Custom Data Types
- 30:45Remove Items from a Set
- 34:30Distributed Data Limitations
- 35:41Use Cases
- 36:38Code Walkthrough
- 43:39Questions
related talks
44mThe Road to Akka Cluster, and Beyond
30mData Consistency Patterns in Cloud Native Applications
55mExperience report on using Akka in production: pitfalls and rewards
43mEffective Eventual Consistency with Actor Models + Amazon Web Services
36mEasy Eventual Consistency with Actor Models + Amazon Web Services
48mThe three paths of Scaling with Akka
01h 00mThe Secret Life of Distributed Systems
38mHow To Build a Distributed System (And Should You?)