Autoscaling is the automatic adjustment of cloud resources based on demand. A system can add more capacity when traffic rises and reduce capacity when demand falls.
What autoscaling uses
Autoscaling policies may use CPU usage, memory, request counts, queue length, schedules, or custom metrics. The goal is to balance performance and cost.
Autoscaling needs monitoring
Autoscaling depends on useful measurements. Poor thresholds can cause slow responses, overprovisioning, or unstable scaling. That is why cloud monitoring matters.
Autoscaling is one of the core advantages of elastic cloud infrastructure.