cancel
Showing results for 
Search instead for 
Did you mean: 

Date as KF in report, and Cube cannot be compressed

Former Member
0 Kudos

Hello All,

Q:

1. I have a report where the Date is used as a Key Figure, and also at the CUBE.

2. Now on deltas all other Key Figure are negated but not the date Key Figure.

3. This doesn't allows me to perform compression for obvious reasons of date as KF.

We face enough performance issues due to complexity of the report and huge no. of records at CUBE, Would appreciate any responses to achieve compression with zero elimination.

I have tried few solutions including formula variable with replacement path- but this requires the base object to be in the result set. Also tried to populate date as DEC(Number) in database and use function process value as date, however it aggregates the values and then convert to date which is sometime 'X' for exceeding the limit.

Also tried with different aggregations of SUM and MAX at info object, doesn't works .

Please post your ideas.

Thanks,

TA

Accepted Solutions (0)

Answers (1)

Answers (1)

yasemin_kilinc
Active Contributor
0 Kudos

Hi Taana,

Just an idea. It is obvious that you use the date as KF for making some calculations in some reports. If the number of formulas in the reports are not too much, then you may try adding each formula as a KF to the infocube and get rid of date kf in the cube. You can fill in these new KFs through transformation rules. (You will need to remodel the infocube to fill in the old data, and make sure that you get a copy of your infocube so if something goes wrong with remodeling then you can go back.)

Regards

Yasemin...

Former Member
0 Kudos

Dear Yasemin,

Thank you for your reply, and I understand your solution, however, the KF calculation at BEx are to achieve dynamic values based on all drill down(most summarized and most granular) hence cant calculate the values at DB.

My flow is DSO--> CUBE,

I tried DEC/INT(KF1) at DSO to get the the before images in changelog on DELTA, and then i mapped this KF1 to KF2, KF2 is DATS on DEC with aggregation as MAx, This approach however doesn't negates the date(KF2) in cube and I am there where I was.

Thanks,

TA