Working with dashboards

The following instructions will create bar and donut charts. Before following these instructions, make sure PosApp is loaded in the Samples workspace, deployed, and running. (See the Hands-on quick tour for an introduction to creating dashboards.)

Once you have added a new visualization to a dashboard, the next step is defining its query. Until you do this, the drop-down field menus in the visualization editor will be empty.

For example, the query for the PosApp donut chart above is:

select count(*) as Count, Status from MerchantActivity group by Status;

This results in four Count values representing the number of merchants with each status. The Count values control the size of the slices and the Status values provide the labels.

By default, dashboard queries are executed only once, when the visualization is loaded. To update the data, refresh the browser to reload the visualization.

To have the a query update continuously, enable polling in the visualization properties. (This property is hidden if the query contains AND PUSH). The query will then be re-run every five seconds, or, if the query takes longer, as soon as it completes.

When querying a WActionStore, the following additional syntax is available:

SELECT ... FROM <WActionStore> 
[
{ <integer> { SECOND | MINUTE | HOUR } |
  JUMPING <integer> { SECOND | MINUTE | HOUR } |
  <integer> { SECOND | MINUTE | HOUR } AND PUSH }
] ...

The SAMPLE BY clause (see CREATE CQ (query) may be useful when a visualization contains too many data points. For example, consider this chart using PosApp data with the query SELECT * FROM MerchantActivity ORDER BY time:CREATE CQ (query)

The dashboard shows as much of the data as it can hold, which is only the last half hour. If you change the query to SELECT * FROM MerchantActivity ORDER BY time SAMPLE BY count, the dashboard can show 3-1/2 hours of data:

To create a drill-down link similar to the one that leads from the Main page to the "Company details" page in PosAppDash:

Return to the source page and click whatever is associated with the drill-down. For example, for a line chart or map, click a point. The drill-down visualization should appear, populated with the data defined by the Source Field value for the point you clicked.

Drill-downs can be used within a single page to filter data interactively. For example, the PosApp visualization below can be filtered by clicking on the pie-chart labels:

If you click the WARM label of the right-hand pie chart, the map and heat map display only data for merchants currently in that category:

This interactivity is defined by setting the drill-down configuration in the pie charts, then using their Id values as in the queries for the two other visualizations. The drill-down configurations are:

Since these visualizations are on the Interactive HeatMap page, the drill-down filters the data without switching to another page.

The US map's query is:

select * from Samples.MerchantActivity [15 minute and push] where (:status 
IS NULL or (Status = :status)) and (:category IS NULL or (Category = :category)) 
group by merchantId;

When the page is first loaded, the :status and :category values are null, so the map displays all data. When you click the WARM label, the :category value is set to WARM, and the map updates accordingly.

To clear the filter and return to viewing all data, click Clear All (next to the funnel icon at left).

To explore this more, run PosApp and go to the Interactive HeatMap page.

Use the Embed command to generate HTML code for an iframe you can use to display a Striim dashboard page in a web page. Embedded dashboard pages include both page-level and visualization-level search and search and filter controls.

The embedded dashboard will not require a Striim login. Instead, it uses a special user account with READ permission on the dashboard's namespace and SELECT permissions on the dashboard, its queries, and related the CQs, streams, and WActionStores.