- It is a service which role is to monitor our apps and automatically adjust capacity to maintain steady and predictable performance at lowest cost
- From a single place we can manage auto scaling of multiple resources across multiple services
- We point to our app and select the services and resources we want to scale (no need to setup alarms and scaling actions for each service)
- To search for resources/services we can use CloudFormation Stacks, Tags or EC2 Auto Scaling Groups
- We build Scaling Plans to automatically add/remove capacity from our resources in real-time as demand changes
- Supports Target Tracking, Step and Scheduled Scaling Policies
- Services supported by Application Auto Scaling:
- AppStream 2.0 Fleets
- Aurora Replicas
- Comprehend endpoints
- DynamoDB Tables and GSI
- ECS Services
- ElastiCache for Redis Replication Groups
- EMR Clusters
- KeySpaces Tables
- Lambda Provisioned Concurrency
- MSK Broker Storage
- Neptune Clusters
- SageMaker Endpoint Variants
- Spot Fleet Requests
- Auto Scaling Groups
- other custom resources