Serverless
Running code without managing servers — the cloud provisions, scales, and bills per-execution, scaling to zero when idle.
Serverless is a cloud model where you deploy code and the provider handles all the infrastructure — provisioning, scaling, patching — running your code on demand and billing only for actual execution time. There are still servers; you just do not manage them, and idle costs nothing (scale to zero).
Worked example: an image-resize function deployed to AWS Lambda runs only when an upload arrives, automatically scaling from zero to thousands of concurrent executions and back, with no capacity planning and no bill when no images are uploaded. Gotcha: serverless trades control for convenience — cold starts add latency, execution time and memory are capped, long-lived connections and state are awkward, and per-request pricing can cost more than a VM at steady high volume — so it shines for spiky, event-driven, stateless workloads, not everything.