Lite Optimization - BPC 7.5 NW

[former_member210696]

In this blog, I will try to discuss lite optimizaion, what it does, how can we see its effect on the infocube at the backend.

Optimization is a process that enhance the system performance. It cleans up data storage which improves the responsiveness of the system. In BPC, there are two types of optimization which can be performed viz. Lite Optimization Full Optimization.

Lite optimization is a maintenance process similar to the BW InfoCube performance maintenance tasks of index rebuild, statistics rebuilds, and compression. Lite optimization is performed on an application within the application set. It basically calls a process chain content of which can be viewed upon in backend. Hit RSPC, under BPC:Example, we can see the process chain corresponding to Light Optimize.

In a nutshell, lite optimization performs following tasks -

1. It compresses/collapses the records.

2. Additional zero suppression is possible (manually adjusted in process chain).

3. It deletes the infocube indexes and moves records from the F to the E fact tables.

4. Updates the infocube statistics.

Let's discuss each one these in detail -

1. Compression of Records-

To view the transaction data corresponding to a particular key, follow below steps in backend BW–
• Enter transaction RSA1 -> Locate Business Planning and Consolidation Infoarea (/CPMB/BPC) -> Locate your appset infoarea -> Expand Infoarea corresponding to your appset
• Locate the infocube corresponding to your BPC application -> Right click and Manage -> On the right side of the screen, click on Contents tab
• Click on Infocube Content. It should display below screen in which select characteristics by clicking on the third button on the screen (the dropdown). Also, select /CPMB/SDATA under key figure right at the end -

 

• Execute and in the next screen, provide the dimension members for each of the dimensions for which you know there exists multiple records in the system. Also, click "Field Selection for Output" and click "Select All". Below is the screenshot for dimension members that I selected in my case -

• After you have made the selection, hit execute and the next screen will show all the records with above key. Aggregated value of the four records can be viewed by selecting the SDATA column and click on Sigma button. Screenshot below -

Before Lite Optimization

As we see, there are four entries for the same record. It’s because that the record was modified four times as indicated by different Request ID in the penultimate column. So, after lite optimization, there will be only one record equal to 430.

After Lite Optimization -

2. Additional zero suppression - Records with transaction value '0' which exist in the system can be supressed using lite optimization. But, for that the process chain will have to be modified to be able to do that. Open the process chain Lite Optimize using RSPC. Right click on the process Collapse -> Display Variant -> Edit the process -> Select the option "Switch on Zero Elimination" at the end as shown below. Save the process variant and come out of it.

If the above change is made and then lite optimization is run, then the records with zero transaction data will be suppressed.

3. Record Movement from F Fact Table to E Table -

F table contains all the numerical values of the transaction data. F table is surrounded by the various dimension tables. Consider this process of moving records from F Fact table to E table as aligned to the compression process discussed earlier. After the optimization, the four records that we see in F table corresponding to our dimension selection will be aggregated and moved to E table and we will not see any record in the F fact table.

Also, in E table the transaction data is stored in system understandable language, so we won’t be able to make out any information by looking at the E table, so in this blog, I will not show the E table per se.

To view content of F table, follow below steps –

• Enter transaction RSA1 -> Locate Business Planning and Consolidation Infoarea (/CPMB/BPC) -> Locate your appset infoarea -> Expand Infoarea corresponding to your appset

• Locate the infocube corresponding to your BPC application -> Right click and Manage -> On the right side of the screen, click on Contents tab

• Click on Fact Table. Here, enter the SIDs for the corresponding dimension members of the record that we viewed earlier i.e. SIDs for below dimension members (in our case) –

• TIME=2013.MAR

• P_ACCT=Revenue

• Category=ACTUAL

• Entity=North

• Product=Dummy_Prod

• RPTCURRENCY=LC

Click Execute after you enter SIDs for the above dimension members and it should display below screen -

Before Lite Optimization -

So, here we see that there are four records in F table.

After Lite Optimization -  Since all the records in F table are aggregated and moved to E table, we don't see any record in the F table now.

4. Updates the infocube statistics -

The infocube statistics are used by the system to optimize the query performance. After larger data load, it is recommended that statistics should be updated for better performance. So, additionally, lite optimization also updates the infocube statistics

For information on full optimization, please refer to the SAP notes 1551154 and 1527747

References -

SAP notes 1508929, 1508929.

http://help.sap.com/saphelp_nw04/helpdata/en/80/1a6473e07211d2acb80000e829fbfe/content.htm