Practical Guide to Modern Networking Telemetry

Chapter 4. Using Telemetry, Individually

As we said at the end of Chapter 3, having covered the ways in which you can manage the data once it’s been collected (in Chapter 2) and the ways in which you can interact with the observability software (in Chapter 3), we now are going to explore detailed use cases and real-world applications. Our hope is that this gives you concrete examples upon which to base your own implementations, so you can put your data to work for you in your environments.

This line of exploration—detailed real-world examples—will carry through for the rest of this book. The difference between Chapters 4, 5, and 6 is that each will expand the circle of our focus, from individual elements, to multiple elements, to (almost) the full application stack.

The Forms of Data

We’ll start with the ways individual data points can help us understand our network, infrastructure, and environment. These items usually take one of four different forms. We’re not describing these simply to be dogmatic. Rather, these descriptions will underpin the use cases that come after.

Metrics

The goal of a metric is to provide a measurement of an element, system, or device at a specific point in time. They’ve been around since the dawn of monitoring and are predicted¹ to outlast all other types of telemetry right until the heat death of the universe.

Metrics are used to tell you everything from whether a system is up or down, whether it’s full (and if not, how much is left), to notifying ...