Getting Started with Streaming Analytics
Also available as:
PDF
loading table of contents...

Chapter 8. Creating Visualizations Using Superset

A business analyst can create a wide array of visualizations to gather insights on streaming data. The platform supports over 30+ visualizations the business analyst can create. For visualization examples, see the Gallery of Superset Visualizations.

The general process for creating and viewing visualizations is as follows:

  1. Whenever you add new data sources to Druid via a Stream App, perform the Refresh Druid Metadata action on the Superset menu.

  2. Using the Superset Stream Insight UI, create one or more "slices". A slice is one business visualization assoicated with a data source (e.g: Druid cube).

  3. Using the Dashboard menu, add the slices to your dashboard and organize their layout.

[Note]Note

Note that when a SAM app streams data to a new cube using the Druid processor, it will about 30 minutes for the cube to appear in Superset. This is because Superset has to wait for the first segment to be created in Druid. After the cube appears, users can analyze the streaming data immediately as it is streaming in.

Creating Insight Slices

The following steps demonstrate a typical flow for creating a slice:

  1. Choose Slices on the Menu.

  2. Click + to create a new Slice.

  3. Select the Druid Data Source that you want to use for the new visualization:

  4. Select a Chart Type from the menu.

    This example creates a "Sunburst" visualization where we are rolling up multiple dimensions like route, eventType and driver info. Configure the chart and click Execute Query.

  5. Another visualization could be integration with MapBox Here we are mapping where violations are occurring the most based on the lat/long location of the event

  6. To save the slice, specify a name and name and click Save.

Adding Insight Slices to a Dashboard

After you create slices, you can organize them into a dashboards:

  1. Click the Dashboard menu item.

  2. Click + to create a new Dashboard.

  3. Configure the dashboard: specify a name and the slices to include in the Dashboard.

  4. Arrange the slices on the dashboard as desired, and then click Save.

Dashboards for the Trucking IOT App

The IOT Trucking app that we implementing using the Stream Builder was streaming violation events, alerts and predictions into three cubes:

  • violation-events-cube

  • alerts-speeding-drivers-cube

  • alerts-violation-predictions-cube

Based on the powerful visualizations that SuperSet offers, you can create the below powerful dashboards in minutes.

IoT Dashboard

Alerts Dashboard