If you are unable to create a new account, please email support@bspsoftware.com

 

News:

MetaManager - Administrative Tools for IBM Cognos
Pricing starting at $2,100
Download Now    Learn More

Main Menu

Differences between Calculated Measure and a Data Item ?

Started by sanchoniathon, 14 Jul 2009 01:27:18 PM

Previous topic - Next topic

sanchoniathon


Hi,

I have these 2 reports which are ALMOST identical.
The first one runs in 5-1o seconds.
Second report runs in 1 minute.

First report uses CALCULATED MEASURES created by us in the query.
Here is an example of what these calculate measures have as EXPRESSIONS:
   tuple([Waste store Amount $];[1])

Second report uses DATA ITEMS created by us in the query.
Here is an example of what these data items have as EXPRESSIONS:
   tuple([Waste store Amount $];[1])

SO THEY HAVE THE SAME EXPRESSIONS ...


Is it possible or does any of you know of any PERFORMANCE issues when using calculated measures VS data items ??
I need to know why one of the report is running much FASTER than the oher ?...


Thanks in advance !

crn.siva

Hi,

When working with dimensional data sources, create Calculated Members or Calculated
Measures where the expression is a member or a tuple-based (as opposed to property-based)
value expression

i think the calculated measure is tuple based thats why the first report runs fastly insted of using data item in second report.

please let me know, if another way.

sanchoniathon

Hello kss001,

- Just a clarification, the first report that is running in 5-10 seconds is using the
  Data Items and the second report, which is taking 1 minute is using Calculated
  Measures.

- Thanks for the information

- We are using a package that is pointing to a BW cube.

Anyhow, i am still not sure why one runs faster as for other reports that uses calculated measures are running fast. Maybe something in conjunction with the tuple and the calculated measure and other thing in the problematic report is making it slow to run ??

Mmore information to come if i have more.

Thanks !