An autoscaler watches demand and adjusts capacity to match it: adding instances as load climbs, removing them as it falls. It is how modern infrastructure stays fast at peak and cheap at rest, with no one awake to press the button.
Why "autoscaler"
Elastic capacity is the default assumption of the cloud. Kubernetes runs pod autoscalers alongside cluster autoscalers that add and remove entire nodes. Every major cloud offers auto scaling groups that grow and shrink fleets against a target metric, and serverless platforms scale from zero to thousands and back without being asked.
The pattern reaches further. Inference clusters scale GPUs with request volume, queue workers scale with backlog, and FinOps teams treat autoscaling as the main lever between reliability and spend. Measure, decide, act, repeat: a control loop with a job title. The word describes the mechanism itself.
Contact
autoscaler.si is for sale, privately. Contact for registration history and transfer terms.
The portfolio
misalignment.si, plus nine names for the layers of modern infrastructure, each offered privately.