Creating a Dataset

Modified on Mon, 27 Aug 2012 06:48 PM by Steve C. — Categorized as: Datasets

Click to return to: Documentation | Defining Datasets


This article shows how to create a dataset in &dw;. A dataset defines the tabular data from a virtual table or virtual cube that you want to visualize on a dashboard.

Creating a dataset from the Home Screen

From the Dundas Dashboard Home Screen, you can create a dataset by clicking Create a Business Metric and choosing the Dataset option. In this case, the name of the dataset will be determined automatically. You can rename the dataset from the Design Explorer at a later time.

The Home Screen.

The Home Screen.

Creating a dataset from the Design Explorer

When you create a dataset from the Design Explorer, you have the option of choosing a name for your dataset and also choosing its destination. For example, if you have a sub-folder in the Design Explorer called My Datasets, you can choose to create a new dataset in that location instead of the top-level Datasets folder.

To create a dataset from the Design Explorer:

  1. Click Design Explorer in the Sidebar.
    Accessing the Design Explorer from the Sidebar.
  2. Select the Datasets folder, then click its menu button.
    The Dataset menu.
  3. From the Datasets menu, choose New Dataset.
  4. In the text box below the Datasets folder, type a name for the dataset, then press ENTER.
    Dataset name.
    The new dataset appears as an item in the Design Explorer and the Dataset Setup Wizard is displayed.

Using the setup wizard

The Dataset Setup Wizard lets you perform the initial configuration of your dataset. The wizard takes you through the steps of selecting a base virtual table (or virtual cube) and then choosing columns for your dataset.

See Using the Dataset Setup Wizard.

When you have completed the steps in the wizard, the Dataset Designer will be displayed in the content area of &dw;. Use the Dataset Designer to perform more detailed configuration of your dataset as shown in the following sections.

The Dataset Designer.

The Dataset Designer.

Configure general settings

Use the General tab of the Dataset Designer to configure general settings for your dataset, all of which are optional:

  1. Click the General tab in the Dataset Designer.
  2. Enter a description for your dataset in the Description box. You can enter text up to 2000 characters in length. The description appears in a tooltip when you hover with the mouse over a dataset item in the Toolbox during the dashboard design stage.
  3. Enter a URL in the Additional info URL box. Use this option to link to a website that provides more information about your dataset. You will be able to access this link from the Additional Information dialog.
  4. The General tab also displays information about the data connector and the virtual table (or virtual cube) that your dataset is based upon.
  5. The minimum polling interval for a dataset is initially set to Database default, which is a database-defined value. To override this, select the Custom option and enter a new value in the seconds box. Use this option to specify your own limit on how frequently the data can be automatically refreshed or updated.
  6. Use the Preferred visualization dropdown list to choose the data visualization control you want the dataset to be rendered as by default (&ie; when you drag the dataset onto a dashboard).
  7. By default, there is no limit placed on the number of data points that can be returned by a dataset. For performance reasons, you can set a limit by unchecking the No limit option and entering a value in the Maximum number of data points box.
  8. Select a single value representation for this dataset which can be used in data rule-based notifications and more.

Define dataset columns

With &dw;, you can define the columns of your dataset by basing them on the columns of a virtual table (or the measures/dimensions of a virtual cube). In the case of a virtual table, you can optionally make use of existing standard or time dimensions, which allow you to add more capabilities to your dataset, such as support for advanced filtering and drill-downs.

See Defining Dataset Columns.

States and state groups

Use the State Groups tab to set the single value representation for your dataset, and to define state groups and their corresponding states.

See Defining States for a Dataset.

Ordering and sorting rules for a dataset

Use the Ordering and Sorting tab in the Dataset Designer to re-order dataset columns and to specify sorting rules (⪚ ascending order, descending order) for the individual values of a dataset column.

See Ordering and Sorting Rules for a Dataset.

Preview and save your changes

Follow these steps to complete the definition of your dataset:

  1. Preview the resulting data from your dataset by using the Preview tab in the Dataset Designer. For more details, see Previewing a Dataset.
  2. Click Save in the Dundas Dashboard Toolbar to save your changes.
    The Save button.
  3. Click Check In from the Dundas Dashboard Toolbar.
    The Check In button.
    Optionally enter a comment in the Check In dialog, then click Check in.

You have created a dataset and previewed its data as a means of verifying its design. You can now use this dataset to drive data visualization controls on a dashboard.

Creating a dataset from a virtual table

If you already have a virtual table, you can create a dataset from it that includes all columns by default:

  1. In the Design Explorer, select a virtual table and do a right-click.
  2. From the menu, choose Create Dataset.
    Create Dataset from virtual table.

A new dataset is created in the Datasets folder with the same name as the virtual table. Rename the dataset as desired and then edit it in order to complete its configuration. By default, all columns from the virtual table are included and made visible.

Related topics

Click to return to: Documentation | Defining Datasets