May 17, 2005 at 1:38 pm
Hi all!
If I have some mdx I'm using in reporting services like this:
select
{ [TimeByMinute].[All TimeByMinute].[2005].[May].[1] : [TimeByMinute].[All TimeByMinute].[2005].[May].[6] } on columns,
{A_list_of_measures } on rows
from ACD_Calls
The column names are unique to the day of month- which means when I use a table to display this in reporting services, the field names change dynamically when the date parameters change which means the table stops working.
I'll post this in reporting services too but I thought maybe I could alias the column names in mdx shielding the reporting services table from changes in dates.
What do you think? Would a matrix be more flexible in this case?
Richard
May 20, 2005 at 8:00 am
This was removed by the editor as SPAM
May 24, 2005 at 2:34 am
May 24, 2005 at 12:31 pm
Thank you, Vadim, for your comments. I would like to try your suggestion but I don't know much about Utility Dimensions. Could you please provide a reference for me to lookup or some other information?
Thanks again,
Richard
May 25, 2005 at 8:50 am
A utility dimension is just a regular dimension like any other. This dimension will have only one regular member. You can create a dimension like this from a table with just one entry (create a table called dimUtility with columns Key and Value , then insert a row with Key=1 and Value='Current' into it and build a regular dimension from it). This is just a design technique for when you need to perform the same calculation(s) on a set of measures. In your case there are no extra calculations to be performed, the sole purpose of it here is to allow retrieval of value for each tuple (Date,Measure) from a set returned by the CrossJoin. You end up with a result set containing date in one column, measure name in another and a column for the actual value for that measure on that date.
You can read more about utility dimensions in a book called MDX Solutions
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply