How to prepare a distributed nodejs application
summary
A Node.js monolith of about 32k LOC and an npm shrinkwrap near 1k lines gets split into services that register via mdns today but fall back to additional engines (SLP work in progress) because of mdns issues on Linux. Services declare hard vs soft dependencies so dependents auto-start and auto-stop, retries and load balancing happen automatically, transactions use a redis (or Kestrel) working-queue for safe retries, remote deps live in package.json, and the system runs on AWS.
AI-written summary · about
chapters (6)
- 0:00Start
- 0:56AUTOMATIC SOFTWARE TESTING
- 1:10SERVER
- 2:46DISTRIBUTED APPLICATIONS
- 12:46STREAMS VS TRANSACTIONS
- 14:09HOW IT WORKS
related talks
01mHow npm split a monolith and lived to tell the tale
44mMicroservices - the lean way (NodeJS)
52mMicroservices the Lean Way NodeJS
40mDeploying node.js at scale
27mProgressive Microservices with NodeJS
48mMicroservices with NodeJS
43mPatterns for Reactive node.js Microservices
23mBuilding Distributed Systems with Node.js
