cancel
Showing results for 
Search instead for 
Did you mean: 

Is this script logic behavior correct?

Former Member
0 Kudos

Hi experts,

we have upgraded to SP5 and we face strange behavior of script logic which we didn't face in SP4.

Currently script logic creates zeroes in every combination of master data regardless if reference data exist or not.

For example: Copying from one month to another with script logic now copies all data which are posted in the from month but also generates tons of zeroes on all the other master data combination.

Performance is suffering a lot and since we can not restrict the values to be processed <> 0 in script logic, this is getting a big issue.

I have written an OSS message for this but would really appreciate any other input.

Regards

Jan

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Hi Jan,

We had a similar problem on a SP04 (with some patches to higher levels).

What we had to do is to is to specify the value of all the other dimensions that we actually wanted to write to. We were lucky in that one of our dimensions had propoerties that pointed to the intersection of some of our other dimensions. so the code looked something like:


*WHEN P_ACCT
  *IS SOURCE
    *REC(FACTOR=1,P_ACCT="RESULT", DESTINATION=CONTRACT.DEST_ID, CUSTOMER=CONTRACT.CUST_ID, ENTITY=CONTRACT.PROFIT_CTR)

Note sure if you have the same setup, but it maybe of help.

Mark

Former Member
0 Kudos

Hi Mark,

thanks for your post. Unfortunately we can nt use this in our scenario since we don't have such a clear dependency between dimension members.

Anyway I find it really strange that script logic acts this way. In big scenarios this is a big performance issue.

I will update this post as soon as I get a relevant answer from SAP.

Regards

Jan

Former Member
0 Kudos

Hi Jan

May I know if you used EXPRESSION in the script logic ?

Halomoan

Answers (1)

Answers (1)

esjewett
Active Contributor
0 Kudos

Jan,

Have you tried turning on compression with zero-elimination on your BPC infocubes? This should get rid of the 0 values and help improve your performance.

Ethan

Former Member
0 Kudos

Hello,

the zeroes are not saved in the cube but are generated during the runtime of the script which critically decreases performance.

This is the sample script we are running:

//=============================================

// Calculation for Opening Balances

//=============================================

*XDIM_MEMBERSET FLOW=BAS(FLOW_TOT)

*XDIM_MEMBERSET INPUTCURRENCY=CZK

*XDIM_MEMBERSET INTCO=INTCO_#, CZ101N, CZ101L

*XDIM_MEMBERSET INVESTEE=INV_#

*XDIM_MEMBERSET LOB=LOB_#

*XDIM_MEMBERSET P_DATASRC=BAS(SRC_TOT_ADJ)

*XDIM_MEMBERSET TEAM=TEAM_#

*XDIM_MEMBERSET P_ACCT=1010211010

//variable

*XDIM_MEMBERSET TIME=2008.12

*XDIM_MEMBERSET CATEGORY=100

*WHEN P_ACCT.TYPE

*IS AST, LEQ

*BEGIN

*REC(FLOW = P_ACCT.OPENFLOW, TIME = 2009.01

,CATEGORY = 100, P_DATASRC = BRIDGE, TEAM = TEAM_#, NOADD)

*END

*ENDWHEN

*COMMIT

Regards

Jan