Serverless

Serverless here means billed-per-invocation or allocated-resource functions and the cases where that model is cheaper or more expensive than a cluster or VM.

Who this topic is for

Architects choosing compute shape for a workload that is already specified.

Problems these pages cover

A long-running worker was placed on functions and now pays for idle concurrency or timeouts.
A cluster is kept for a handful of cron jobs.
The comparison never included operational surface: deploys, networking and observability.

Guides in this topic

Ankit Mehta Updated Aug 30

When serverless fits a production platform

Serverless

Teams pick Lambda, Azure Functions, Cloud Run or Function Compute because the slide said “no servers,” then discover a timeout, a VPC ENI delay, or a concurrency quota during the first peak. The problem is matching the unit of work to a product whose limits you have actually read.

Related runbooks

Autoscaling failure

Consulting

Scoped review for this topic: Cloud architecture consulting.