Distributed Systems explained with NodeJS
summary
Shows a Node.js implementation of a replicated key–value store and runs three concrete variants across four local nodes (ports 3001–3004): two‑phase commit (CP/AC style), a sloppy‑quorum/Dynamo style (AP), and a CP leader/quorum implementation. Demonstrates that two‑phase commit blocks or times out when a replica is slow or down, sloppy writes return immediately and rely on read‑repair to converge, and the CP variant uses leader election to maintain consistency; network partitions are simulated with iptables and code/examples are published on GitHub.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 3:13Why distributed systems
- 10:02Scalability and partitioning
- 16:14CAP Theorem explained
- 21:56Two-Phase Commit demo
- 33:24AP vs CP systems demo
- 40:55Conclusion & Q&A
related talks
23mBuilding Distributed Systems with Node.js
01h 13mConsensus Algorithms. What Does Node.js Have to Do with It?
29mWhy would we ever build a distributed computing platform in node?
45mDo Distributed Systems they said. It'll be fun they said.
25mGrowing Distributed Systems
40mImplementing Distributed Consensus
23mLightning Talks on Distributed Systems
44mDistributed systems and scalability rules