Charts enable you to visualize and perform sophisticated analytics on the metrics you are sending in to SignalFx. At the end of this tutorial, you will understand how to find the metric you want to plot, how to format a basic chart, and how to save it to a dashboard.
Before you start
Before you can create charts in SignalFx, you need to be sending in metrics. In this tutorial, you’ll be using sample data that we are sending in for you. Note: This tutorial is based on sample data that is being sent into trials created after May 28, 2015. If your trial or organization was created before that, then this does not apply.
Open up the chart builder
To get started, make sure you are in your home page by clicking on the SignalFx logo in the upper left hand corner. From there, select the dashboard named 'Your first dashboard' from the sub navigation menu.
As you haven’t added anything yet, you will see a big gray button in the middle of your Workspace. Click on that:
You can also always click on the Add Chart button in the upper right hand corner of any dashboard.
You should be in the chart builder view.
Choose the transaction count metric
Let’s start by choosing the metric that you want to plot. We are sending in three metrics to your organization:
In this first chart, we’ll use demo.trans.count in our example.
Start by clicking on Browse next to the signal field. In the Find a Metric or Time Series lightbox, you will see metrics and time series that are available in your organization. In the search bar, type demo.trans.count. In the list of search results, you should find a line that says demo.trans.count. Click on that line to select it (making sure that the box to the left of the demo.trans.count text is checked), then click ADD to add the metric to your chart.
You should have one or more plots appearing above in the plot window.
Display plot as a stacked area chart
With SignalFx, you have a number of choices as to how you want to display your plot. Depending on the type of chart, you may want to use area charts, bar charts, line charts, lists or heatmaps. In this example, we’ll use a area chart.
Hover over the chart display selector and make sure Graph is selected. Next, hover over the chart type selector and choose Area.
Your plot should now be appearing as a set of overlapping area graphs. To see it stacked (such that you can see the values of each time series and how they contribute to the total), go to the chart Settings menu and check the box for Stacked Chart.
Change the time range
Depending on the metric, you may want to change the time range that appears in the chart. The default for new charts is 15 minutes; if your time series have a higher, or finer resolution (new datapoints arriving every few seconds) then you may want to make the time range smaller. Conversely, if your time series has a lower, or coarser resolutions (new datapoints arriving every few minutes, as in the case of Amazon Web Services CloudWatch metrics), then you will want to make the time range larger, so as to fit more datapoints within the window.
To change the time range, go to the time picker in the top right of the page. The default of 15 minutes appears here, as -15m. Change the time range to 5 minutes by clicking into the time picker and typing -5m.
Smoothing out the chart
Charts can sometimes appear to be jagged because of the nature of the data. If you are interested primarily in the broader trends within the data, rather than the exact value of each data point, it can be useful to apply a smoothing function.
In SignalFx, we do this by applying an analytics function that will transform the data, using a moving average with a window of 30 seconds. To do this, click on the stylized f (for “function”) the analytics field and select mean, then select mean:transformation. This will bring up a window that allows you to choose the size of your moving average window; enter 30 for the value and choose seconds as the unit.
Viewing the legend
Now move your mouse cursor into the plot window. A legend appears with information about the plot line that your cursor is hovering over.
Click on the chart to open a detailed legend under the chart window. Each line in the legend table represents one plot line. The legend will show you the value of each plot at the time point that your mouse is currently hovering over (as "value"), the values at the time point that you clicked on (as "pinnedValue"), and a number of other properties associated with those values.
Annotate your chart
Now that you have the metric displaying properly, it’s time to annotate it so that your chart will be understandable by other people who see it. First, hover over the chart settings menu again and add a label to Y Axis 1 by typing Transactions into the Label field.
Next, make sure that the y-axis of your chart starts from zero by typing 0 into the Min field immediately below the label. Depending on the metric you’re plotting, this may or may not be desirable.
Finally, add a chart name and description. Click into the Chart Name field and type something concise and descriptive, like ‘No. transactions’, then click into the Chart description field to provide more information.
Save your chart
To save, click on the Save and Close button in the upper right hand corner. You will be returned to the dashboard that you started from, and should now see the chart that you created. Note that you can resize charts by dragging from the lower right hand corner of individual charts, or move them by dragging from their upper right hand corner.
Congratulations! You’ve created your first chart.