Learning OpenTelemetry

Chapter 9. Rolling Out Observability

Just because the standard provides a cliff in front of you, you are not necessarily required to jump off it.

Norman Diamond

Telemetry is not observability, as we’ve said in previous chapters. It’s a necessary part of observability, but it’s not sufficient in itself. So if telemetry isn’t enough, what are the other factors you should consider when rolling out observability to your organization, team, or project? This chapter answers that question.

We’ve written this chapter with a broad audience in mind—not just site reliability engineers (SREs) or developers, not just engineering managers or directors. The true value of observability lies in its ability to transform organizations, and to provide a shared language and understanding around how software performance translates into business health. Observability is a value, in the same way that trust and transparency are values. Observability is a commitment to building teams, organizations, and software systems in ways that allow you to interpret, analyze, and question their results so you can build better teams, organizations, and software systems.

This task is not for any one individual or group. It requires an organizational commitment, bottom to top, around how to use data as an input to processes, practices, and decision making. To that end, this chapter presents several case studies of organizations and projects that have implemented OpenTelemetry and uses them to present roadmaps that ...