9 Measuring your platforms

 

This chapter covers

  • Learning the importance of measuring platform performance
  • Implementing DORA metrics and learning the secret continuous improvement
  • Using tools and standards to collect and calculate metrics

In chapter 8, we covered the principles of how to build a platform that helps you deliver software and enables your teams to have the tools they need when needed. This last chapter is all about making sure that the platform is working, not only for application development teams, but for the entire organization. To understand how the platform is performing, we need to be able to measure it. There are different ways of taking measurements on the software we run. Still, in this chapter, we will focus on the DORA (DevOps Research and Assessment) metrics, which provide a good foundation for understanding our organization’s software delivery speed and how good we are at recovering from failures when they happen.

This chapter is divided into two main sections:

  • What to measure: DORA metrics and high-performant teams
  • How to measure our platform initiatives:
    • CloudEvents and CDEvents to the rescue
    • Keptn Lifecycle Toolkit

Let’s get started by understanding what we should be measuring, and for that, we will need to look at the DORA metrics.

9.1 What to measure: DORA metrics and high-performant teams

9.1.1 The integration problem

9.2 How to measure our platform: CloudEvents and CDEvents

9.2.1 CloudEvents for continuous delivery: CDEvents

9.2.2 Building a CloudEvents-based metrics collection pipeline

9.2.3 Data collection from event sources

9.2.4 Knative Eventing event sources

9.2.5 Data transformation to CDEvents

9.2.6 Metrics calculation

9.2.7 Working example

9.3 Keptn Lifecycle Toolkit

9.4 What’s next on the platform engineering journey?

9.5 Final thoughts

Summary