BW 3.5 Inventory Loads and Aggregates

[Former Member]

I am currenlty working on a reload of our inventory data into our production system. I have the help documentation for loading but we have added aggregates to our cubes for specific reports. Can anyone tell me when to load the aggregates, before or after compression and/or marker update?

In the current data we did the initial after compression and marker update but in the process chain it is running prior to compression and marker update.

Also curious, how this effects the data if you do it one way verses the other?

Thank you for your prompt reply.

Caroline

[Former Member · Accepted solution]

Hi,

first load the aggregates and then do the compression. Compression is done on aggregates to improve query performance. Therefore you should first load your aggregates.

Compression

http://help.sap.com/saphelp_bw30b/helpdata/en/ca/aa6437e7a4080ee10000009b38f842/content.htm

Aggregate

http://help.sap.com/saphelp_nw04/helpdata/en/c5/40813b680c250fe10000000a114084/frameset.htm

http://help.sap.com/saphelp_nw04/helpdata/en/4f/187d3bce09c874e10000000a11402f/frameset.htm

Forum

Regards

Andreas

[Former Member · Accepted solution]

Thank you for the information. When loading the inventory analysis the loads fail if you do the aggregate load first. There is a specific way that it has to be done and when aggregates are involved it does something completely different.

I was truly curious about how it effects the marker update as we have had issues with this and that is why we are reloading. I believe the marker update is a seperate record number so i believe it should go up fine, but I wanted to be sure.

Thanks again and if you know anything regarding the marker update I would be glad for the info 🐵

Caroline

[Former Member · Accepted solution]

Hi,

I have found a good how to paper regarding Inventory loads and marker update:

[How To…Handle Inventory Management Scenarios in BW|https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/com.sap.km.cm.docs/library/how-to-guides-rig-official/netweaver-2004/business-intelligence/how%20to%20handle%20inventory%20management%20scenarios%20in%20bw%20(NW2004).pdf]

Hope this helps.

Regards

Andreas

Edited by: Andreas Schindler on Apr 24, 2008 8:25 AM

Edited by: Andreas Schindler on Apr 24, 2008 8:26 AM

[Former Member · Accepted solution]

thanks, I have this document and I don't believe it discusses anything about marker update and aggregates. Again thank you for the info.

Caroline

[Former Member]

Thank you!

[Former Member]

Hi Caroline,

as long as I know the aggregation cube can be load after you do the compression and marker update.

If you do first aggregation cube then when do compression and marker update it will not work because the concept of compression is remove the compression data from F Fact table into E Fact Table (you do aggregate means the data is exist on both F Fact Table and E Fact Table)

I Hope this help you.

Cheer

Ir

[Former Member]

Hi Corline,

In your case you can load the Initial Stock 'BX' with the marker, this will bring the Stock on hand.

And immediately you can fill your aggregates(i mean Rollup the data then only the data will available for reporting.)

Run the Material Movements 'BF' wiht no marker, this will bring you the material movements untill that point from the begining.

You can run fill the aggregates for this load as well (Rollup).

Then with the Marker in place you can run the Deltas for Material Movements this will be the regular load in the Proces chain.

In your process chain after this Delta Load the Roll step should be there.

Note:- Befor Reinitializing delete the data From Targets and the Aggregates as well.

To delete the Aggregate Data Deactivate all the aggregates on that cube.

Hope this will help you........!!

Best Regards,

VNK.

[Former Member]

I am good with how to load inventory and you would think that they way you have things written down is how it works but it doesn't, we have tried this load a few times and for some reason the aggregate loads cause failures. thank you so much for the info, much appriciated.

Caroline