We already know what the Continuous Delivery process is and why we use it. In this section, we describe how to implement it.

Let's start by emphasizing that each phase in the traditional delivery process is important. Otherwise, it would never have been created in the first place. No one wants to deliver software without testing it first! The role of the UAT phase is to detect bugs and to ensure that what developers created is what the customer wanted. The same applies to the operations team—the software must be configured, deployed to the production, and monitored. That's out of the question. So, how do we automate the process so that we preserve all the phases? That is the role of the automated deployment pipeline, which consists of three stages as presented in the following diagram:

The automated deployment pipeline is a sequence of scripts that is executed after every code change committed to the repository. If the process is successful, it ends up with the deployment to the production environment.

Each step corresponds to a phase in the traditional delivery process as follows:

• Continuous Integration: This checks to make sure that the code written by different developers integrates together
• Automated Acceptance Testing: This replaces the manual QA phase and checks if the features implemented by developers meet the client's requirements
• Configuration Management: This replaces the manual operations phase-configures the environment and deploys the software

Let's take a deeper look at each phase to understand what is its responsibility and what steps it includes.

The Continuous Integration phase provides the first feedback to developers. It checks out the code from the repository, compiles it, runs unit tests, and verifies the code quality. If any step fails, the pipeline execution is stopped and the first thing the developers should do is fix the Continuous Integration build. The essential aspect of the phase is time; it must be executed in a timely manner. For example, if this phase took an hour to complete then the developers would commit the code faster, which would result in the constantly failing pipeline.

The Continuous Integration pipeline is usually the starting point. Setting it up is simple because everything is done within the development team and no agreement with the QA and operations teams is necessary.

The automated acceptance testing phase is a suite of tests written together with the client (and QAs) that is supposed to replace the manual UAT stage. It acts as a quality gate to decide whether a product is ready for the release or not. If any of the acceptance tests fail, then the pipeline execution is stopped and no further steps are run. It prevents movement to the Configuration Management phase and therefore the release.

The whole idea of automating the acceptance phase is to build the quality into the product instead of verifying it later. In other words, when a developer completes the implementation, the software is delivered already together with acceptance tests which verify that the software is what the client wanted. That is a large shift in thinking about testing software. There is no longer a single person (or team) who approves the release, but everything depends on passing the acceptance test suite. That is why creating this phase is usually the most difficult part of the Continuous Delivery process. It requires a close cooperation with the client and creating tests at the beginning (not at the end) of the process.

There is usually a lot of confusion about the types of tests and their place in the Continuous Delivery process. It's also often unclear how to automate each type, what should be the coverage, and what should be the role of the QA team in the whole development process. Let's clarify it using the Agile testing matrix and the testing pyramid.

Brian Marick, in a series of his blog posts, made a classification of software tests in a form of the so-called agile testing matrix. It places tests in two dimensions: business or technology facing and support programmers or critique the product. Let's have a look at that classification:

Let's comment briefly on each type of test:

• Acceptance Testing (automated): These are tests that represent functional requirements seen from the business perspective. They are written in the form of stories or examples by clients and developers to agree on how the software should work.
• Unit Testing (automated): These are tests that help developers to provide the high-quality software and minimize the number of bugs.
• Exploratory Testing (manual): This is the manual black-box testing, which tries to break or improve the system.
• Non-functional Testing (automated): These are tests that represent system properties related to the performance, scalability, security, and so on.

This classification answers one of the most important questions about the Continuous Delivery process: what is the role of a QA in the process?

Manual QAs perform the exploratory testing, so they play with the system, try to break it, ask questions, think about improvements. Automation QAs help with nonfunctional and acceptance testing, for example, they write code to support load testing. In general, QAs don't have their special place in the delivery process, but rather a role in the development team.

You may look at the classification and wonder why you see no integration tests there. Where are they up to Brian Marick and where to put them in the Continuous Delivery pipeline?

To explain it well, we first need to mention that the meaning of an integration test differs depending on the context. For (micro) service architecture, they usually mean exactly the same as the acceptance testing, as services are small and need nothing more than unit and acceptance tests. If you build a modular application, then by integration tests we usually mean component tests that bind multiple modules (but not the whole application) and test them together. In that case, integration tests place themselves somewhere between acceptance and unit tests. They are written in a similar way as acceptance tests, but are usually more technical and require mocking not only external services, but also internal modules. Integration tests, similar to unit tests, represent the "code" point of view, while acceptance tests represent the "user" point of view. Concerning the Continuous Delivery pipeline, integration tests are simply implemented as a separate phase in the process.

The previous section explained what each test type represents in the process, but mentioned nothing about how many tests we should develop. So, what should be the code coverage in case of unit testing? What about acceptance testing?

To answer these questions, Mike Cohn, in his book Succeeding with Agile: Software Development Using Scrum, created a so-called testing pyramid. Let's look at the diagram to understand it well.

When we move up the pyramid, the tests become slower and more expensive to create. They often require touching user interface and hiring a separate test automation team. That is why acceptance tests should not target 100% coverage. On the contrary, they should be feature-oriented and verify only selected test scenarios. Otherwise, we would spend a fortune on the test development and maintenance, and our Continuous Delivery pipeline build would take ages to execute.

The case is different at the bottom of the pyramid. Unit tests are cheap and fast, so we should strive for 100% code coverage. They are written by developers and providing them should be a standard procedure for any mature team.

I hope that the agile testing matrix and the testing pyramid clarified the role and the importance of acceptance testing.

Let's move to the last phase of the Continuous Delivery process, configuration management.

The configuration management phase is responsible for tracking and controlling changes in the software and its environment. It concerns taking care of preparing and installing the necessary tools, scaling the number of service instances and their distribution, infrastructure inventory, and all tasks related to the application deployment.

Configuration management is a solution to the problems posed by manually deploying and configuring applications on the production. Such common practice results in an issue whereby we no longer knows where each service is running and with what properties. Configuration management tools (such as Ansible, Chef, or Puppet) enable storing configuration files in the version control system and tracking every change that was made on the production servers.

An additional effort to replace manual tasks of the operations team is to take care of application monitoring. That is usually done by streaming logs and metrics of the running systems to a common dashboard, which is monitored by developers (or the DevOps team, as explained in the next section).