Aggregation creates higher-level data from lower-level data by calculating the values up a hierarchical dimension. The most common method of aggregation is summarization, but many other methods are available, including hierarchical and weighted methods.
User-defined aggregation rules identify how and when the base-level workspace data, which has been acquired from another source, is rolled into the parent members of the dimension hierarchies.
Analytic Workspace Manager enables you to specify aggregation rules in these ways:
On the cube: The cube has default aggregation rules that apply to all of its measures. You define these rules as part of the definition of the cube.
In a cube script: A cube script can provide additional rules for one or more measures within a cube.
Aggregation is typically performed immediately following a data refresh.
You can execute cube scripts using the Maintenance Wizard. The aggregated values overwrite any values that were previously calculated as part of a data refresh.
A default aggregation plan enables you to define the aggregation rules once and use them for all measures in a cube.
To use a default aggregation plan, take these steps:
Create a cube and complete the Aggregation tab.
On the navigation tree, right-click the cube and choose Maintain Cube from the pop-up menu.
Follow the steps of the wizard to load and aggregate the cube.
A cube script enables you to control additional aggregation parameters and to specify the order in which the measures are aggregated.
To create an aggregation step in a cube script:
On the General tab of the cube script, click New Step, then select New Aggregation Step. The New Aggregation Step property pages are displayed.
Complete all tabs. On the Status page, select the base level for regular composites. Select all levels for compressed composites. Always aggregate data from the base level up. If the base level does not have data values, then allocate down to the base first.
Click Help for additional information about each tab.
Save the aggregation step, then save the cube script.
To execute the cube script, right-click the cube script in the navigation tree and choose Execute Cube Script.
View the aggregated data in the measure viewer.
Copyright © 2003, 2007, Oracle. All rights reserved.