Creating Your First Dashboard

In the previous chapters, we've mostly concentrated on panels and how to use them and configure them. We did this pretty much exclusively on the dashboard, which is the canvas that we display our panels on. In the next few chapters, we will zoom out from the panel level to the dashboard level. We'll continue to learn more about various panels, but this will mostly be in the context of making our dashboard layouts and queries more efficient.

In this chapter, we're going to take on the task of designing a couple of dashboards – one packed with information suitable for viewing on a workstation or laptop, and another containing only key pieces of information suitable for being viewed at a glance or from a distance.

In both cases, we'll pick up some workflow techniques that can help speed up the often laborious task of creating, configuring, and laying out panels...

Technical requirements

The tutorial code, dashboards, and other helpful files for this chapter can be found in this book's GitHub repository at https://github.com/PacktPublishing/Learn-Grafana-7.0/tree/master/Chapter07.

Designing a dashboard

Before we get started and work on a new dashboard, it's best to have a plan of action. Ask yourself a few questions:

• What information do I want to convey?
• What is the visual context for the dashboard?
• What is most important; what is least important?

Let's take a look at these questions in more detail.

Conveying information

In the case or our dashboard, we will be building a dashboard that can be used to produce a forecast of the upcoming weather. For this purpose, will need to describe the following conditions:

• Current temperature and dew point
• Barometer reading and trend – rising, falling, or steady
• Wind direction and speed
• Visibility

We also want to know the current temperature as that will help us decide what to wear, for example. The dew point is an indication of humidity (and relative comfort, depending...

Creating a high information density dashboard

In our first example, we'll be constructing a fairly detailed dashboard of graph panels. This dashboard is similar to the one you might find accompanying a metrics-driven server application. It's intended to provide a number of metric graphs that might also serve as the top layer for further drill-down exploration. In our example, we'll be assembling a series of graphs to cover the weather metrics we've scraped from the National Weather Service using the application we developed in Chapter 5, Visualizing Data in the Graph Panel.

If you've been following my instructions in the previous chapters, you're probably well aware of how much work goes into getting a panel just so, and you can be forgiven for being a little anxious about the idea of creating a lot of panels for a dashboard. It's a valid concern, and for these examples, I'm going to take you step-by-step, as well as offer some effort...

Creating a high information visibility dashboard

In this second example, we'll create a dashboard intended to provide information in a much higher level of view; that is, that of a display intended to be scanned rapidly in order for us to get a big-picture viewpoint. Typically, you'd see this type of dashboard in a kiosk-type context, such as in an operations center or a public informational display.

We'll be conveying a set of data that's similar to what we had in the previous section, but we'll only have a limited slice of each data series, typically the most current readings. We'll be making extensive use of the Stat panel as opposed to the Graph panel, as we did previously. The idea we're trying to convey is that the dashboard will be displayed in a context that makes details hard to read from a distance.

Designing the dashboard

What we want to do is create and arrange a set of panels that will fit on a single page...

Summary

That completes our exercise of working with dashboard creation and layout. Play with your dashboards' panel arrangements to see what various combinations look like. This is a good opportunity to get a better understanding of how to work with the Grafana layout manager. While you experiment with the ordering of the various panels, keep a few things in mind.

First, cultural groups can read from left to right, right to left, and from top to bottom. Know your audience and arrange your dashboard panels to reveal information in the order that your viewers typically scan. Second, use color, size, and visual contrast to draw the eye of the viewer toward the information you want to particularly highlight. Finally, depending on the context, you may want to avoid packing too much information onto a single dashboard. Too much visual information can be confusing to the viewer.

In the next chapter, we'll look at more ways to make panel creation more efficient...