How CloudFormation works

The CloudFormation service reads all its input commands straight from the template. The template is stored in an AWS-managed S3 bucket, and a link to the template is provided to us when we deploy the stack. By default, the template is read in parallel, meaning that the service will try and create all the objects at once. It is very important to use the DependsOn attribute if any services need to be deployed in sequence. The following steps will be taken in CloudFormation when creating and managing an environment:

1. The user writes or visually designs a specification document called a template.
2. Using the management console, the CLI, the SDKs, or the CloudFormation API directly, the user sends the template to the CloudFormation engine.
3. The CloudFormation engine processes the template and verifies its syntax.
4. The CloudFormation engine deploys the resources...