TM1 Forum

Hi!

I have something like 15 worksheets report that reference one reporting cube, but different slices, all figures are pulled by DBRW.
Currently I'm trying to optimise performance of the whole report.
I have figured out that I need something like three View() functions to cover required slices.

Should I place all 3 View() functions on the title Worksheet and reference them from other sheets or place one View() function on each of 15 sheets?

Thanks.

The difference will be small, if different at all. But I would prefer to have 1 formula in each of the sheets. Personal preference.

I would always go with appropriate VIEW formulas per sheet as per the data that is being retrieved. (depending on number of cubes referenced or different slices from the same cube this could mean multiple view formulas per sheet).

This covers you if only recalculating a single worksheet and not the whole excel application (which is what users should be trained to do anyway is they are working with TM1). I'm not sure how exactly excel handles worksheet calculation priority order when doing the application.calculate event but this would also cover against the possibility of a cell being queries before the view formula is passed and the stargate view created. Also it is just neater and easier to troubleshoot is the view formulas are sitting in the worksheets that need them.

Thanks, will make one View function per sheet.

Hi guys,

Sorry to hijack this thread but I'm looking for an answer that is closely related to this question.

In a workbook with multiple slices, it's possible to get each slice (a tab in the workbook) working by either using a regular formula such as
=VIEW("Servername:CUBENAME","!",$B$2,"!","!","!",$B$3,"!")
Where there are 7 dimensions in the cube and dimensions 2 and 6 are in the title area (and are therefore listed with a hard coded reference) and the other dimensions are either in the row or the column (and are referenced with a "!"). The question is, I can get the slice to work by either using the VIEW formula as previously shown or in a more simple form such as:
="Servername:CUBENAME"

If both options work as they appear to do, what is the disadvantage of using the second option where the dimension references are omitted?
I ask this not because I want to deliberately break functionality that IBM have created and which works, but because I've discovered a number of the latter 'solutions' being used at a clients site and am wondering what impact it might have on the system and if it's worth changing all these formulas in the slices.
Many thanks.

Using the VIEW command invokes the Stargate cacheing technology which significantly improves retrieval speed on anything of any size.

Thanks David, glad to know it doesn't affect any functionality or workbook/calculation integrity. I'll implement the view as better performance is always welcome!