- Manages a fleet of EC2 instances, automatically adjusting the count based on demand
- Scales out (adds instances) during high load, and scales in (removes instances) when demand drops
- Maintains a minimum and maximum number of instances at all times
- Replaces unhealthy instances automatically
- Can integrate with Elastic Load Balancer to register/deregister instances automatically
- ASG service is free — you pay only for the EC2 and other underlying resources
Key Capacity Settings
- Minimum capacity – The lowest number of instances to keep running
- Desired capacity – The target number of instances based on current needs
- Maximum capacity – The upper limit of instances allowed
Benefits
- Elasticity – Matches compute capacity to demand
- Fault tolerance – Replaces failed instances automatically
- Cost optimization – Scales in during low demand to reduce cost