Auto scaling is a cloud feature that automatically increases or decreases the number of running instances or containers in response to load, based on rules or metrics such as CPU use, memory or requests per second.
Why it matters
It keeps performance steady during traffic spikes and reduces cost when demand falls. It also helps recover from instance failures, since an unhealthy instance can be replaced automatically.
Scaling is not instant. New instances need time to boot and warm up, so rules must anticipate load. It also depends on the application being stateless or sharing state externally.
Example
A rule says: add two instances when average CPU stays above 70% for five minutes; remove one when it stays below 30% for fifteen minutes. A minimum of two and a maximum of ten instances keeps costs bounded.
Common mistakes
- Having no maximum limit, so a bug or attack scales costs sharply.
- Storing sessions or uploads on the local disk of an instance that may disappear.
- Using thresholds so tight that the group scales up and down repeatedly.
Related
Get notified when this term updates
Plus occasional practical tips on uptime and monitoring, nothing else. Unsubscribe anytime.
This page is general information written by our editorial team and updated when documentation changes. Servers and providers differ, so test changes safely first. See our editorial policy and disclaimer.