Skip to Content

Working with Semantic Partitioned Cube in SAP BI 7.3

Semantically partitioned object is an InfoProvider that consists of several InfoCubes or DataStore objects with the same structure. Semantic partitioning is a property of the InfoProvider. We specify this property when creating the InfoProvider. Semantic partitioning divides the InfoProvider into several small, equally sized units (partitions).

Main advantage of SPO is to manage mass data with better performance.

Below Topics will be covered.

·         How to create SPO.

·        How to transport SPO.

·        How to use in Multiprovider .

1. To create a semantic partitioned cube please follow the below steps :

1.       Enter T-Code RSA11

2.       Select any particular infoarea where you need to create the Semantic Cube

3.       Create Infocube as shown below.

SS1.jpg

1.       Provide the Technical Name for the SPOJ

SS2.jpg

1.       2. Tick the Semantically Partitioned Check Box in the ‘Property’ Tab as shown below :

SS3.jpg

1.     3.Then click the create button.

2.       This will take you to the Semantic Partition Wizard.

In this wizard, we have modeling section through which we could maintain the Semantically Partitioned Cube. Please see screenshot below

SS4.jpg

1.       4. Create the Dimensions and add the necessary infoobjects under these dimensions similar to the way we do it for a Basic Infocube.

2.       Once the SPOJ is modeled, click on the ‘SAVE’ Button shown below:

SS5.jpg

1.       5. Once the SAVE button is clicked then only the Start Activation Button will work. Otherwise the SPOJ will not be changed. Please refer screenshot below :

SS6.jpg

1.   In the above screenshot also please note the Button ‘Change Partitions’. Here we specify on what criteria we are dividing the cube and into how many partitions.

Please see screenshot below:

SS7.jpg

1.  6.Then click on ‘Change Transformation” Button to modify the mappings between Semantic cube and below Data Target to which it is mapped.

SS8.jpg

1.   7.Click on Manage Data Transfer Processes Button to activate all the DTPs :

Please note that we need to create a template before we proceed with creating DTPs

SS9.jpg

1.  7. Select all the DTPs and click on ‘Generate” Button :

SS10.jpg

1.    We can create the Process chains by clicking on the ‘Create Process Chains’ Button but it’s not mandatory. We can create the Process Chain in normal way by adding all the DTPs.

Transporting Semantically Partitioned Cube :

1. Transporting an SPOJ is a quite complex task and below is the exact sequence we need to follow while transporting :

1.   2. Go to ‘Transport Connection” in Modeling Workbench

2.    3. Select SPOJ and Click on Select Button as below :

SS11.jpg

1.   4.Select the particular SPOJ you need to Transport and click on Transfer Selections as shown below :

SS12.jpg

1.   Then select ‘Only Necessary Objects’ under ‘Grouping’ and ‘List” under ‘Display’ Section.

2.    We need to collect only the below objects which are ticked.

SS13.jpg

Also, collect only the below transformation in the same TR as below :

SS14.jpg

So, when we look into the TR, it should have only below objects and nothing else:

SS15.jpg

Please note that this is very crucial and we are not supposed to collect and transport Partitioned Cubes in any case as it leads to Transport Failures!!!

The SPOJ TR should have only the above objects as shown in the screenshot.

1.    DTPs can be collected and transported separately as below :

SS16.jpg

How to Use SPO’s in multiporviders

1.   There are 2 ways in which multiproviders can be built on top of the underlying Semantic Partitioned Cubes:

Scenario 1 :

Include separately each and every semantically partitioned cube without including the Parent SPOJ. Please see screenshot below :

Here disadvantage is we need to add manually every time we create new partition so maintenance will be more.

SS17.jpg

Scenario 2:

Add only the main SPOJ in the Sem.Part.Prov. as below , this will give you advantage on less maintenance as in future if you add new partition it will be considered default here.

SS18.jpg

To report this post you need to login first.

9 Comments

You must be Logged on to comment or reply to a post.

  1. mukesh kumar

    Hi Madav,

    Good Document.In this document Madhav has explaine about collecting the TR’s also.

    Keeping posting this type of document. 🙂

    Regards,

    Mukesh

    (0) 
      1. Pallavi Pande

        Few queries

        How does this work with the existing Info cube which is Partitioned?

        I have cube partitioned till 2006 to 2011.

        I have cube with huge data now and I need to partition it with respect to Month as suggested by SAP in early watch report…

        Can I create SPO in this case?

        (0) 

Leave a Reply