Parameters, mappings & outputs

Make a template reusable.

Hard-coded templates are single-use; Parameters make one template serve many deployments. A parameter declares a typed input with an optional default, allowed values, and constraints, and you supply it at deploy time. The program refers to it with !Ref, so the same template can deploy a t3.small in dev and a t3.large in prod purely by changing the parameter.

Parameters:
  Environment:
    Type: String
    AllowedValues: [dev, staging, prod]
    Default: dev
  InstanceType:
    Type: String
    Default: t3.small
    AllowedValues: [t3.small, t3.large]
 
Resources:
  AppServer:
    Type: AWS::EC2::Instance
    Properties:
      InstanceType: !Ref InstanceType
      Tags: [{ Key: Env, Value: !Ref Environment }]

Mappings and Outputs

Mappings are static lookup tables resolved with !FindInMap — the classic use is picking a region-specific AMI. Outputs publish values from the stack: an endpoint, a bucket name, a security group ID. An Output can be exported with a name, which lets another stack import it (cross-stack references, covered later) — making Outputs both a human-facing result and a machine-facing contract between stacks.

Mappings:
  RegionAmi:
    us-east-1: { ami: ami-0abc123 }
    eu-west-1: { ami: ami-0def456 }
 
Resources:
  AppServer:
    Type: AWS::EC2::Instance
    Properties:
      ImageId: !FindInMap [RegionAmi, !Ref "AWS::Region", ami]
 
Outputs:
  ServerId:
    Value: !Ref AppServer
    Export: { Name: !Sub "${Environment}-app-server-id" }