Does hierarchical filters built using parent-child dimension and using tabular values differ in performance? RRS feed

  • Question

  • Hi,

    I have built a hierarchical filter using Tabular values template and multi-select tree option. I referred this article for building the filter (http://office.microsoft.com/en-us/performancepoint/HA102617851033.aspx). The parent-child table that i have built is according to this article(http://msdn.microsoft.com/en-us/library/aa198072(SQL.80).aspx).

    I am wondering if by using a SSAS parent-child dimension instead of tabular values method helps in decreasing the time that the filter takes to hit the table and display the filter values.

    Let me explain it in a better way. My aim here is to build a hiererchical filter in PPS dashboard which follows the multi-select tree template. If i build a parent-child dimension in SSAS (i am not interested in any aggregates or measures in the cube. In fact, my fact table is an empty table i.e truncated. The fact table is just for the purpose of building a cube. I am interested only in Dimension for building the filter) would the cube have already built parent-child hierarchical structure from the table as it does for aggregates and measure groups. And in turn would it reduce the time in displaying the filter items as the cube has already (my assumption!) built the parent-child hierarchical structure from the underlying table. I think my question would be better understood if the article for building the parent-child table(given above) is seen, i have built exactly in that manner.

    Please do let me know if i have not articulated my question properly and i'll try to explain it in a better way as i am very new to SSAS concepts.

    Thanks and Regards,



    Thursday, April 29, 2010 5:40 AM