How to Build a Serverless Death Star
summary
Claims that a brittle “Serverless Death Star” comes from too many tiny functions, intense coupling and single-person language ownership, leading to cascading failures when one function fails. Demonstrates common anti-patterns: avoid function-to-function sync calls because waiting increases cost and fragility, limit languages to two or three so knowledge isn’t siloed, and reduce heavy libraries, network‑intensive functions and undocumented APIs. Recommends cataloging functions, documenting APIs during development and using asynchronous invocations only for true fire‑and‑forget batch work.
AI-written summary · about
related talks
42mHelp! I've created a serverless monolith!
51mConfusion In The Land Of The Serverless
41mYou Don’t Want to Build a Serverless Monolith
55mConfusion In The Land Of The Serverless
28m8 Keys for Successful Serverless Architectures
01h 00mMicroservices and Serverless
49mYou Don't Want Serverless
52mConfusion In The Land Of The Serverless