In this chapter, we explored the concept of containers using Docker and ECS. After exploring the basics of how Docker works, we created a container for our application. After running locally, we created a new set of resources to run Docker containers on AWS. We did that using the DevOps best practices. We used CloudFormation to generate our resources, treating our infrastructure as code. This allows us to keep those changes under source control. Resource-wise, we created an ecr repository to manage the different revisions of our containers, two ECS clusters with auto scaling capability for staging and production, two ALBs to proxy the traffic to our containers, a set of tasks, and an ECS service to configure and deploy our application.

In the end, we reimplemented a CI/CD pipeline. We did that by using CodeBuild, CodePipeline, and their integration with CloudFormation.

In Chapter 7, Monitoring and Alerting, we will continue improving our systems and implement one of the last key characteristics of DevOps—measuring everything. By taking advantage of a number of features present in the different services we use and by coupling that with other AWS services such as CloudWatch, we will be able to implement a monitoring strategy for our infrastructure and services.