Auto-scaling Group
Noun · Development
Definitions
Auto-scaling Group (ASG) is a cloud infrastructure feature that automatically adjusts the number of compute instances (virtual machines or containers) in a group based on current demand, defined policies, or scheduled events. When traffic or CPU usage increases beyond a threshold, the ASG launches additional instances to handle the load. When demand decreases, it terminates excess instances to reduce costs. Auto-scaling is a core feature of cloud platforms: AWS Auto Scaling Groups, Google Cloud Managed Instance Groups, and Azure Virtual Machine Scale Sets all implement this pattern. Configuration includes minimum and maximum instance counts, scaling policies (CPU-based, request-count-based, custom metrics), cooldown periods (to prevent rapid oscillation), and health checks that replace unhealthy instances automatically. Auto-scaling enables applications to handle variable workloads without manual intervention, paying only for the capacity actually needed. Effective auto-scaling requires stateless application design (so any instance can handle any request) and careful tuning of scaling thresholds and warm-up times to balance responsiveness with cost.
In plain English: A system that automatically adds more servers when traffic spikes and removes them when traffic drops.
Example: "The auto-scaling group adds instances when CPU hits 70% and scales down after traffic drops — we saved 40% on compute costs."