Observability Engineering, 2nd Edition

Chapter 25. The Observability Landscape Through a Systems Lens

In the previous chapters, we established the why and what of accelerating organizational learning speed through feedback loops. In this chapter, we get into the how.

Amplifying feedback loops can spiral into virtuous cycles or death spirals. Balancing feedback loops can create stability or devolve into chaotic swings or rigidity. What makes the difference between learning and flailing is how you’re using those loops, and the quality of the observability tools feeding data into them.

We’ll start by looking at how feedback loops typically work in most organizations. Then we’ll contrast them with the feedback loops that actually accelerate learning, and examine why closing that gap has been hard for so many teams. We’ll look at two generations of observability tooling and which loops each one supports, and we’ll close by examining how AI is both accelerating the need for change and opening up new ways of working.

This chapter focuses on recognizing the capabilities you need so that you can choose observability tooling that fits. The next few chapters build on that by examining the business case, how to diagnose your current investments, and how to drive organizational change.

The Landscape Feels Noisy Because the Labels Are Noisy

Before we begin, let’s talk about the elephant in the room: the “observability” category has become almost unfathomably large. It started as monitoring, logs, and application performance monitoring ...