Help! I've created a serverless monolith!
summary
Explains concrete signs of a serverless monolith and exact remedies: use Durable Functions for orchestration and fan‑out/fan‑in instead of brittle HTTP chains, move shared storage/configuration when splitting function apps, use deployment slots or queues to avoid deploy restarts, and measure MTTR (example: 12 hours/6 incidents = 2h). Recommends Dev Containers / GitHub Codespaces, tweak Application Insights and function scale limit (consumption can hit 200 instances) and shows ably.pizza.dev as a working visualizer for Durable Function workflows.
AI-written summary · about
chapters (10)
- 0:00Intro
- 2:09Serverless monolith
- 6:08Agility
- 9:02MTTR
- 13:27Deployments
- 15:47Complexities
- 18:12Moving Complexity
- 20:54Durable Functions
- 25:04Dual Functions
- 35:02Scalability
related talks
41mYou Don’t Want to Build a Serverless Monolith
35mChopping the monolith
51mConfusion In The Land Of The Serverless
45mThe Evolution From Monolith to Microservices to Serverless and Beyond
09mHow to Build a Serverless Death Star
01h 00mMicroservices and Serverless
05mA Serverless Monitoring Checklist
28m8 Keys for Successful Serverless Architectures