Create a New Data Cube

This applies to: Managed Dashboards, Managed Reports

Create a new data cube to organize your data from one source, or integrate data from multiple sources, starting from a blank cube or by using an available input transform to create your reusable data model. Once you've created and checked in a data cube, you and your users can create content in dashboards and reports that use this reusable data model.

To create a cube, you'll connect content from your data sources as tables, SQL or MDX statements, transforms such as Data Input, Python Data Generator, and the R Data Generator. Use and configure joins, unions, lookups, or fusing to bring your data together, then review the process result.

Create a new data cube

1. Open the data cube canvas. You can do this in one of several ways:

   • Select Create Data Cubes from the home page

   • Select Data from the Managed Dashboards and Reports main menu, then select Create and then select from Blank, Data Input, Python Data Generator, or R Generator

   • Right-click Data Cubes in the Explore window, then select New Data Cube from the menu

2. Add a data structure to the data cube work area. Drag and drop a structure, such as a spreadsheet or table, from a data connector in the Explore window to the canvas, or select an option from the Add/Edit toolbar.

   Note: Data structures in your Symphony environment also include data sources discovered by Symphony in Managed Dashboard and Managed Reports data connectors, as well as shared from Data Discovery data connectors.

3. After you've added a first data structure, Symphony adds a connected Process Result to create a simple ETL process. The data cube is named based on the source of your first data structure: select the name in the data cube status bar to rename it.

4. Continue adding what you need: data structures and joins, scripts, and other transforms. Configure how they work together to complete your data cube structure. See Adding, connecting, and configuring more data structures and transforms.

5. Optionally, use the toolbar to perform tasks from the available menus (Essential, Common, Settings, Storage, Add/Edit). The options available in the Contextual menu may vary based on what is selected on the canvas.

6. Configure your Process Result. Configure the output of the data users with read access see when using it as a data source in metric sets.

7. When you've added, connected, and configured all elements, your options include:

   • Select Check In from the toolbar to make your data cube available to for use in your project and sharing with other users.

   • Use the data cube in a metric set or dashboard before you check it in and make it available to others.

Adding, connecting, and configuring more data structures and transforms

1. When you add a new data structure, transform, or join, it may display on the canvas as a red icon until:

   • You connect it to your ETL process.

   • You configure the connected transform.

   • Configure a join type, set up the unique key binding, or select and deselect columns for inclusion in the output.

   Note: You can select a specific color for a structure on the canvas by changing it in the Configure dialog.

2. Insert or connect a new transform by one of several methods:

   • Select an existing connection between two existing transforms to insert the new one.

   • Select an existing transform and drag it to a transform to connect the two.

3. Configure your newly added transforms. Select the transform, then Configure in the toolbar, double-click the transform, or select Configure from the menu when you right-click on the transform.

4. When you've configured all elements, right-click the Process Result element to configure the final output of the data users with read access. See Configure a Process Result.

Configure a Process Result

1. When you've configured all elements, right-click the Process Result element to configure the final output of the data users with read access. The Data cube elements work area opens.

   

2. Customize each element's name, description, predefined formatting, and more. Any data not used as a measure becomes a hierarchy when output from the cube whether or not it was based on a hierarchy you defined ahead of time to contain multiple levels or other customizations. If not replaced with a predefined hierarchy, a column of data becomes an 'implicit' hierarchy.

3. To replace an implicit hierarchy with your own that's based on matching key data, select the option Select a hierarchy or level to use as a replacement. In the Open dialog that appears, select the hierarchy, or expand it and select a particular level that matches the column's values. Select the Open button at the bottom to proceed.

   

The steps are the same to replace a DateTime type column (such as OrderDate in this example) with a hierarchy from the built-in time dimension or from one of your own, to group by and drill down between levels such as Year and Month.

See Process Result for more information.

Data Cube Options

When you create a data cube, Symphony provides several options to get you started adding your first input transform. If you don't have the appropriate application privileges, some options may be hidden from you.

• Blank. Build a data cube on a blank canvas in the data cube work area.

• Manual Select. Enter a SQL or MDX statement to make a selection from a data connector instead of dragging structures onto the canvas.

• Data Input. Reference a warehouse data storage are containing user input directly, or through a data input interaction.

• Python Data Generator. Generate data by writing scripts using the Python programming language.

• R Data Generator. Generate output by running R scripts against the R server.

Each of these options available to you can also be added as additional input transforms to existing data cubes. See Edit a Data Cube.

Note: You need to be a user with a Developer seat to create or edit a data cube.

For more information, see:

• Data Cubes

• The Managed Dashboards Data Model