Auto Scaling

Auto scaling is a cloud feature that automatically adds or removes computing resources, such as servers or containers, based on demand. When traffic rises, for example during a sale, more instances start; when it falls, extra instances are stopped so you do not pay for idle capacity.

Scaling is triggered by metrics such as CPU usage, request count or queue length, or by a schedule. Horizontal scaling (scaling out) adds more machines, while vertical scaling (scaling up) gives one machine more CPU and memory. AWS Auto Scaling groups, Azure Virtual Machine Scale Sets and the Kubernetes Horizontal Pod Autoscaler are common examples.