Basic Information
| Field | Details |
|---|---|
| English name | Amazon EC2 Auto Scaling |
| Full name | Amazon EC2 Auto Scaling |
| Chinese description | EC2 自动扩缩 |
| Japanese description | EC2 オートスケーリング |
| Exam frequency | ⭐⭐⭐⭐⭐ |
| Often confused with | Elastic Load Balancing / Vertical Scaling |
In one sentence
An Auto Scaling Group maintains EC2 capacity and health; ELB manages traffic, while ASG manages capacity.
Core purpose
- Minimum, Desired, and Maximum define the lower bound, current target, and scaling ceiling.
- A Launch Template defines instance configuration, and an ASG can launch across AZs and replace unhealthy instances.
- CloudWatch metrics, alarms, or schedules trigger a Scaling Policy that adjusts Desired Capacity.
Exam focus
- Target Tracking keeps a metric near a target; Step Scaling changes capacity by alarm range.
- Scheduled Scaling suits predictable peaks; Manual Scaling directly changes desired capacity.
- A common highly available pattern is ALB + Multi-AZ ASG + EC2.
- Before scale-in, consider connection draining, warm-up, local state, and downstream limits.
Common misconceptions
- Auto Scaling does not distribute requests, and ELB does not decide instance count.
- Desired Capacity can change dynamically; it is not permanently fixed.
- A high Maximum does not guarantee databases, quotas, or budgets can support that scale.
Key takeaway
Min ≤ Desired ≤ Max; Launch Template defines instances, and Scaling Policy adjusts capacity.
Related services
Elastic Load Balancing, CloudWatch, Launch Template, EC2, SQS.