The first service we need to introduce is CodePipeline, which is the orchestrator service for the pipeline process. It provides a workflow capability to implement a CI/CD automation pipeline by managing the inputs and outputs of other AWS services. It also provides a way to visualize the entire release process and status with its own management console. The console also does a great job of pulling together all of the resources that are used in the pipeline, such as CodeBuild projects and CodeCommit repositories:

CodePipeline pipelines can be triggered by supported source control systems or uploads to S3. A trigger will kick off the process for building, testing, and deploying if you have configured your pipeline to do that. An added feature is that you can also include approval gates in the process, which will send an email out to the person who approves it, prompting them to approve or decline the next stage. This means you can add a manual intervention step, which will help you change management when deploying to production.

Within a pipeline, we need to be able to run some compute tasks. We will cover this in the next section.