Skip to Content

Creating Semantically Partitioned Object(SPO) in BW 7.3

Applies to:

SAP NetWeaver Business Warehouse 7.30 (BW7.30)

Summary

This paper provides details and step by step procedure for creating Semantically Partitioned Object (SPO) based on DSO.

Company: L&T Infotech Ltd

Created on:  02 November 2012

Author Bio

Vishnudas Narayana Mangalam is working as SAP-BI Consultant in L&T Infotech Ltd and has worked on BW Implementation/Support Projects.

Table of Contents

Introduction

Features Advantages

Reduced Manual Effort

Step by Step Procedure For Creating SPO based on DSO

Related Content

Introduction

A semantically partitioned object is an Info Provider that consists of several Info Cubes or Data Store objects with the same structure. The goal of semantic partitioned objects (SPOs) is to model a semantic partition once and then apply the model repeatedly in order to create and maintain highly scalable data containers while reducing development efforts.

Features

·         An SPO is based on either a DSO or an InfoCube.

·         Each SPO is made of multiple data providers of the same type with identical structure.

·         Each of the underlying data providers behaves like a “regular” DSO or Cube, with its own metadata tables, data tables, data flows and sources/targets, and are independent of each other.

·         Each SPO will have a so called Master provider, which acts as a template based on which the part providers are built.

·         While defining metadata for the SPO, the definition is being made implicitly to the master provider, which is then used as a template to create the   part providers

·         The part providers are read only, changes to them can be made only via the master provider

·         When a SPO is created, a reference structure keeps track of the partitions.  The structure is placed in the Multi Provider for querying.

·         SPO Wizards create all Data Transfer Processes (DTP), transformations, filters for each data store and a process chain automatically.

Advantages

·         Applicable for DSO and Cubes

·         Better performance with mass data

o        Reduced Query Run time for large data volume

·        Close data connection

o        Better Error handling and partitioned data available for reporting and analysis

·          Working with different time zones

o        Separation can be done on time zones. Helpful when we have multiple source systems and single BW instance

Reduced Manual Effort

In earlier SAP NetWeaver BW releases, multiple manual steps are required to achieve semantic partitioning and scalable data containers. For example, if an administrator needs to create 20 partitions (broken down by geography, year, business unit, and so on), he or she has to run through every step manually, linking transformations and data transfer processes (DTPs) and creating process chains. And if the partitions need to be modified, then the administrator has to start over and apply each change manually.

With SAP NetWeaver BW 7.3, semantic partitioning has been significantly simplified with a wizard that automates the creation of SPOs and guides users through the process. An administrator can now use automated semantic partitioning to create InfoProviders that consist of multiple DataStore Objects or InfoCubes with the same structure. This leads to highly scalable data containers when data is loaded. An administrator can also load the objects in parallel, enabling better performance.

This automation drastically reduces the manual effort required to create, maintain, and change semantic partitions, and it enables fast implementation of enterprise data warehousing with SAP NetWeaver BW.

Step by Step Procedure For Creating SPO Based On DSO

Step 1:

Go to transaction RSA1.

From the required InfoArea right click and select the option “Create DataStore Object”.  

1.JPG 

Step 2:

Fill the required Technical details for DSO. Check Semantically Partitioned.          

You can choose a template or add objects manually. SPO can be used as a template for creating another object.

To create a new template Select ObjectType as “Semantically  Partitioned Object” and then click on Create.

2.JPG

To create based on existing template Select ObjectType as “Datastore Object” and then click on Create.

2_1.JPG

Step 3 :

Click Maintain object and in Object Maintenance screen Click on Save.

If any changes are required in maintenance of object, make the necessary changes and save.

3.JPG

Step 4 :

Click on Maintain Partitions.

First step is to choose the infoObject(s) on which the partition criteria is to be defined. This is based on the business logic.

4.JPG

Step 5 :

Select the required characteristics and transfer to selected characteristics list in Partitioning Criteria.

5.JPG

5_1.JPG

Step  6 :

Click on Continue.

Step  7:

Partition 01 is created by default as shown below. Specify the partition criteria.

For giving range uncheck the single value checkbox.

7.JPG

Step 8:

There is an option to add multiple partitions, either based on time or a fixed number of partitions.

The time-based partitioning option allows you to specify number of partitions based on available time characteristics

Add another set of partitions if multiple partitions are required. To avoid creation of partitions manually incase of fixed intervals, click multiple partitions

8.JPG

Step 9 :

In  Partitions Maintenance screen give the necessary details as shown. This depends on interval and partitions we want to create.

9.JPG

Step 10 :

Partitions are automatically created based on the periods per partition. If required these can be edited.

10.JPG

Step  11 :

Once partitions are created Click on Start Activation.

11.JPG

Step  12 :

Once the activation is done logs are displayed as shown below.

12.JPG

12_1.JPG

Step  13:

Click on Create transformation.

Fill the necessary details in the transformation maintenance screen.

1 transformation is sufficient for all the partitions.

13.JPG

The field mappings are automatically generated.

13_1.JPG

Step 14 :

Activate the transformation.

14.JPG

14_1.JPG

Step 15 :

Create Data Transfer Process .

From DTP Templates section create a new template and save the entries.

15.JPG

15_1.JPG

Step  16 :

Select the template and all the datasources in various partitions. Click on Assign DTP’s to selection.

16.JPG

Click on Save

16_1.JPG

Select all the DTPs and click on Generate

16_2.JPG

1 by 1 all the DTPs will get generated automatically.

16_3.JPG

16_4.JPG

Step  17:

Click on Create Process Chain.

17.JPG

Click on Add to add the DTPs

17_1.JPG

17_2.JPG

The sequence of the DTPs can be changed and then save.

17_3.JPG

Click on Generate to activate the chain.

17_4.JPG

Save the start process if the start criteria is as desired.

17_5.JPG

Process chain is now created and activated.

17_6.JPG

17_7.JPG

Step 18  :

The complete data flow for the Semantically Partitioned Object (for DSO ) is as shown below.

18.JPG

Refresh the tree. The entire dataflow is created.

18_1.JPG

Related Content

http://www.sdn.sap.com/irj/scn/weblogs?blog=/pub/wlg/21334Reference 2

https://weblogs.sdn.sap.com/pub/wlg/24825

http://www.sdn.sap.com/irj/scn/weblogs?blog=/pub/wlg/24829

http://scn.sap.com/docs/DOC-8719

To report this post you need to login first.

12 Comments

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

  1. vinod bokade

    Nice article for step by step creation of SPO on DSO.

    Qu1 : Is it possible to perform above on existing DSO, or we need to create a new DSO and move date from old to new ?

    Qu2 : What about multiple fiscal year variant data ? If data in DSO is having 2 fiscal year variants, then how SPO can be handled?

    Thank-You in advance.

    (0) 
    1. George Holdsworth

      vinod:

      In step 7 above, you can either place a single value into a partition definition or multiple.  So, in theory you should be able to put multiple fiscal variants into that box to define the partition. 

      OR,

      You could create a whole new partition for the second fiscal variant by clicking the +PARTITION button and defining it there.  I’ve played with this (using different fields than fiscal variant) and it works great.

      George

      (0) 
      1. vinod bokade

        Thanks George for sharing your experiance.

        I remember we faced issue with earlier version i.e. BI 7.01 in order to use partioning we have to have single fiscvar, or else we need to split the data into diff cube and then apply partioning on each one.

        What about my first qu 1, please share if you have any experiance with that too.

        Thank-You.

        (0) 
        1. Pallavi Pande

          Hi..

          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) 
  2. Daniel Ray

    we have a dso to infocube flow. do we need to create spo only on the infocube right ? No need to do SPO on the dso ?

    2nd question, we need to build a MP on the infocube. Do we need to build the MP after the spo partition or does not matter ? the Multiprovider will automatically get converted from the existing infocube if SPO is built on it ? Please update.

    thanks

    DR

    (0) 
  3. Hello PAMARTHI

    Hi Vishnudas Narayana Mangalam,

    It was very good Document for SPO and it is very interesting to read and understand is well…

    After going through this document, i got a doubt like for example if there is data from 2000 to 2015 and we created a multiple partition to load each year data separately, and if we required to load next coming year do we required to create a new partition for the  upcoming year??

    Thanks & Regards.

    P Sriram.

    (0) 
    1. VISHAL RAINA

      Hi Pamarthi

      you can maintain the existing SPO…by changes the filter conditions. i would say you should create s separate partition for this.

      Regards

      Vishal

      (0) 
  4. Jalina Vincent

    Hi Vishnu,

    Thanks for sharing wonderful document.

    I am trying to create New template of SPO. While creating I am getting the below error as reference structure does not contain any characteristics. I have included the key fields and data fields in the DSO. But still I am getting the error. Can you pls help.

    Some more doubts :

    1. Is it mandatory to create New template for SPO ?

    2. Can you please provide steps to add new fields in all partitions of SPO.

    Thanks.

    (0) 
    1. Sivaramakrishnan M

      Hi Vincent,

      May I know that you had already fixed the issue, if it is so then i can ask you one question about the SPO, suppose if there is a requirement that you need to create a multiple and number of SPo’s if there any FM to do this and any program ??

      Thanks’

      Siva

      (0) 

Leave a Reply