Designing Measure Groups and Measures in C#

Drawer Denso QR Bar Code in C# Designing Measure Groups and Measures

In the previous procedure, you confirmed that Subcategory is the granularity attribute for the Product Forecast measure group. You can see here that there is a direct attribute relationship with Category, so you know that you can use the Category attribute to view aggregated measures for that measure group as well. Are there any indirect relationships 14. Expand the Category target attribute. Category has no direct attribute relationships defined. Therefore, you will only be able to use two attributes Subcategory and Category when working with Product Forecast measures. 15. Switch to the browser in the Cube Designer, remove Subcategory from the grid, and then drag Category to the rows axis. The measures from each measure group aggregate properly because they are directly or indirectly related to the granularity attribute of each measure group. 16. Now remove Category from the grid, and then drag Color to the rows axis. Your screen looks like this:

It looks like something is wrong. The Sales Amount Forecast for each color is the same amount. What has really happened is that, because Color is not directly or indirectly related to Subcategory, the cube is unable to aggregate this measure by color and instead displays the total Sales Amount Forecast for each color. The relationships defined in the DSV are used to build the attribute relationships in the dimension. You can remove these attribute relationships if you prefer, and you can create new attribute relationships if necessary. Be sure to test your changes to ensure the

Part II:

resulting aggregated values are what you expected. Try this out by removing the relationship between Subcategory and Category. 17. Switch to the Dimension Designer for Product, and then in the Dimension Structure tab, right-click the Category source attribute beneath the Subcategory attribute, click Delete, and then click OK to confirm the deletion. 18. Deploy the project. Notice the error message that the Category attribute is not related (directly or indirectly) to the Product key attribute. Removing the attribute relationship between Category and Subcategory left Category as an orphaned attribute. Because it s not related to the key attribute of the dimension directly or indirectly, the dimension cannot be processed. 19. Close the Error List dialog box. 20. Drag the Category attribute to the space labeled <New Attribute Relationship> beneath Product. Now Category is directly related to Product, but has no relationship with Subcategory. The dimension can now be processed, but what will happen when you view measures with the Category attribute 21. Deploy the project, switch to the Cube Designer, click the Reconnect button on the toolbar of the Browser tab, drag Color off of the grid, and then drag Category to the grid on rows. With Category, you see the same behavior with Sales Amount Forecast that you saw previously when Color was on rows. Even though the DSV has a join between Subcategory and Category, it is the attribute relationship in the dimension structure that determines whether or not aggregated values can be derived for an attribute that is not the granularity attribute for a measure group. 22. Now that you ve seen how attribute relationships affect the aggregated values of measures, restore the attribute relationships in the Product dimension to the original state by removing the Category source attribute from the Product target attribute and placing it under the Subcategory target attribute. 23. Switch to the Dimension Designer for Product, drag Category from beneath Product, and drop it in the space labeled <New Attribute Relationship> beneath Subcategory. The next time that you browse the cube, after the project is deployed, you ll see the correct aggregated values for Category across all measure groups.

5: