DevOps Overview
DevOps is the combination of development and operations teams working together toward a unified goal: Ship the highest-quality code and infrastructure in the shortest span of time to deliver value to customers faster. With DevOps, operations are a core part of every step in the development pipeline. This includes the developers writing the code as well as the engineering teams that provision the hosting infrastructure and build and manage the release pipeline. It includes release engineers, database operations, network operations, security operations, and many others. The Microsoft model for DevOps shows how teams go through four phases, as shown in Figure 6.1.
• Plan: Use a backlog that defines a 1- to 3-week sprint, and define a prioritized list of user stories. For an ecommerce app, an example of a user story would be that the product catalog microservice enables a customer to browse for items grouped by category.
• Develop and Test: Next, develop and test the user story, which includes creating automated unit, performance, and integration tests, which are discussed later in this chapter.
• Release: When the new version is ready to release, use an automated process for deployment to make it repeatable and reliable.
• Monitor + Learn: Diagnostics and monitoring after a release enables the team to understand how customers use your services and monitor service health.
Whereas Chapter 7, “Monitoring,” discusses monitoring and diagnostics in detail, the key here is to remember that you aren’t “done” once you deploy your release.