Software → System Design
Autoscaling
Automatically adjusting the number or size of running resources in response to demand.
Motivation
Autoscaling solves the problem of changing capacity automatically as demand changes.
Where it fits
It belongs to system design and operations-aware architecture.
Mental model
The system watches signals such as CPU, queue length, request rate, or latency, then adds or removes capacity.
Important details
Good autoscaling needs safe thresholds, cooldowns, health checks, and capacity limits.
Common mistakes
- Autoscaling without understanding bottlenecks.
- Scaling too aggressively and causing instability.