Master data modelling optimum way

Hi guys,
I am new to SAP BW and I would really appreciate your valuable advice on a subject. I have the following scenario:
There are three entities (banks) that have business partners (clients), approximately 1-2 million each. I would like to store that data in BW. One way would be to use 0BPartner which is the BI Content ready solution of SAP to store all of them using a compound key to differentiate them. The other would be to create three custom info-objects eg. XBPartner, YBPartner and ZBPartner and use each to store the corresponding Business Partner of each bank accordingly. Each one has the same fields and comes from 0BPartner template. The reason to do that would be to avoid having such a huge info-object. Is this a valid reason?
Secondly if I wanted to have a report using data from all three custom info-objects which way would be the best? Maybe an info-set or a multi-cube? Would the entries of the three tables be outer joined this way (like we want them to) or would there be a problem in reporting?
Any help would be kindly appreciated.
Points will be awarded
Regards
Yannis
Edited by: Ioannis Dimopoulos on Jan 3, 2008 8:25 AM

Thank you sendmeinfo,
Keeping in one info-object all clients from all three banks would mean that I need a way to differentiate between them. There is the option to use a compounding key for example KEY1 = Client# and KEY2 = Bank Code using those keys. There is also the other option to use one concatenated key that derives from the two previous keys KEY1KEY2 joining those two keys together. Which way do you think is best?
Thanks a lot
Yannis

Similar Messages

Master Data modeling with data from several source clients

Hi people,
We have a modeling issue at hand..
From several R/3 sources, separated as Clients (eg. client 100, 101, 102 etc.) we will load data (transactional and Master Data) to one BW system.
Now, I am wondering what is the best way to avoid Master Data issues here, when data can be different for the same objects in the various source systems (clients)
Example:
In source client 100, 0MATERIAL 1000 = PLASTIC
In source client 101, 0MATERIAL 1000 = WOOD
I have tried to understand how compounding with 0SOURCESYSTEM works, but I can't see how that will work for us, since we may have several clients on the same sourcesystem (e.g. client 100 and 101 can be on the same system). This would not help me separate the masterdata.
- Does anyone have experience with this they could share?
- Is it a possibility to add 0LOGSYS or something to our InfoObjects that we will load?
- Do we then add the InfoObject to the cube in the end?
Any inputs that can clarify this is appreciated. Points awarded ofcourse!
Cheers,
Jon A.

Hi,
Donot go with one server to define it as a source system .I mean ,Even though you have only one R/3 server . you have 3 clients in it. So go with Clients to define the source systems.
So in BIW You must have 3 different source systems in Source system tree. Am I irght?
So all of them has different logical system names.
So use 0LOGSYS as compounding characterstic.
With rgds,
Anil Kumar Sharma .P

Export Master Data Model from MDM solution.

Hello,
I have been assigned the requirement to integrate our product with your “Master Data Management” solution. To be more specific, I need to import the Mater Data structure (model), from your companies MDM solution, into our product to build its table/column structure. What I am looking for is technical documentation that will help me make this possible.
Here is a list of some questions I have that technical documentation could address:
•     Does your MDM solution have the capability to export the underlying Master Data structure (model)?
•     What form is the data structure exported (db schema, data dictionary, etc)?
•     What format would the exported information be in (XML, delimited text file, SQL DDL)?
•     If the format is XML, is there an XML Schema that defines what it will look like?
•     Do you have any sample export files that I can work with?
•     Is there technical documentation that would help me make this possible?

If you want to integrate with DRM I would suggest using the API (check out the SDK). This is the preferred and supported way to inject or extract from DRM from an integration standpoint. You can also do imports/exports, for example exporting a hierarchy in XML, to a table, etc.
Directly modifying the tables is not supported or encouraged; the DRM engine enforces many rules like validations, keeps multiple versions, tracks changes in transaction logs, etc, besides the fact that the database structure can change during upgrades. Additionally, due to the dynamic configuration capabilities some things like property values are stored in a one-to-many table that may be pivoted from what you expect.

Upload Master Data GL quicker way instead of using FS00?

Hi Guys
I would like some advice regarding uploading Master data into SAP
As part of my training course we are having to upload live data onto SAP and our trainer has suggested using SM35 and to click new recording then typing FS00 in transaction code and then we put our details for one transaction and then the recording should suffice.
I do not want to manually type them all in using FS00 even though it is only 40 GLs as I want to practice LSMW
Also I have been told of a another way by typing in SA38 then RFBISA00. I have tried this however getting error message Message no. BL001
I know this may sound stupid but the Excel file is saved in a text format do I keep the headers and the rows where it says required or not. Or do I just copy the raw data?
Thanks guys
MD

Hello,
To get best answers, you will have to post the question in ABAP Data Transfer Forum.
For LSMW, you may be refer the link [LSMW|http://www.****************/Tutorials/LSMW/LSMWMainPage.htm]
This will provide details step by step document on how to use LSMW.
Regards,
Abhishek

Central Master Data Management

Hi All,
Can anyone tell me about central master data management ?
In what way it is difference from consolidation and harmonization?
I have gone through several documents but i did n't get any idea on that topic.
If anyone have knowledge on CMDM, Please explain in their own words.
Thanks
Narendra

Consolidation and Harmonization procedures are very different from CMDM (central master data management), however often times CMDM is not very useful without considering the other two as well.
Consolidation essentially means creating identity around your master data across your landscape. More specifically you want to identify duplicate records and merge them into one record. Harmonization is the process of pushing the new cleansed data back out to your partner systems.
Central master data management is the process of creating and managing enterprise level attributes in one place. One common mistake that people make is that central master data management means having one place to create your records and containing ALL data. This is not correct, MDM focuses on enterprise level attributes. In other words, which attributes are KPI's, shared across multiple processes in scope, shared across multiple systems, important for reporting, need ultimate quality, etc. Once you have determined your data model then you begin developing the workflow (if needed) around the creation of master data. This way you can easily keep your data clean and free of duplicates moving forward following a consolidation and harmonization process.

Error while assigning infoobject to the read master data in rule type

Hi sdns,
I want to read the zwostd info object data from zemployee master infoobject attribute list. For that i have selected rule details of zwostd infoobject in transformation of info cube, then i am tring to select the read master data rule type. It is taking more time and then it displayed error like permitted time was exceeded. Could any one please give me the solution for this problem.
thanks
karun

Sometimes directly typing the infoobject name in the field helps - otherwise it tries to fetch all the infoobjects available which is possibly why the time out occurs..?
Enter the Master Data infoobject directly and then you will get the fields assigned.
However my experience with read master data has been that whenever you change the master data IOBJ by way of adding / removing Nav Attributes / attributes etc - all the transformations using the read master data for the IOBJ that is being changed will get deactivated and it is a pain reactivating and transporting them again - particularly so when you are developing something on them and the changes are stuck across multiple requests ...
IMHO use a routine instead to do the same - it involves lesser maintenance. SAP is anyway going to run a similar SQL...

Issue with Vendor Master Data-field notes ID (Table ADRT-REMARK)

Hello,
I have a issue related to Vendor Master Data maintenance.
The client wants to maintain the u201CNotesu201D field in vendor master data. The field gets populated from the field u201CADTR-REMARKu201D table ADRT.
Now client wants to maintain some alphabet to distinguish some of vendors (i.e. 1250) out of 5000 vendors. We tried to develop LSMW, however it is not getting generated as this field is not getting recorded as the field is inside the tab (next to e-mail id field) which requires click. It seems BDC also is also not working as it is not populating these fields except e-mail id.
The table ADRT contains filed ADDRNUMBER which is the same field in LFA1, and it is ADRNR (Address number).
Also please note that CATT/e CATT is not permitted at all the client level.
Can any body suggest the way to change the master data through some way.
Regards,
Chepan

Hi,
Did you heard about SAP GUI Scripting? There is SAP functionality - seems to Excel Macros. You are able to record SAP like macros, then after small code modifcation (code is in Visual Basic), and special excel origin (there could be excel prepared like for CATT) upload data to VMD.
If you have any questions I try to help you?
Best regards,
Piotr
Edited by: Piotr Wieczerzak on Jan 26, 2010 9:35 PM

Best book or online education on SQL developer data modeler 3.0 version

Hi,
i dont see any OBE for sql developer data modeler the way we have for sql developer, would one of you please suggest the Best book that explains everything about data modeler tool or any other online tutorial for that matter, I am new to this URL and been asked to work on ER (Logical), relational, physical models extensvily.. right away, i will have to build or starts righht from ER/logica and ahead or reverse, forward enginerring. thanks for help friends.

Did you look at Data Modeler web page.
There are some online demos and documentation.
Regards,
Ivan Zahariev

Data model in Interactive Reporting

Hi,
Could any one clarify me on What is Data model in Interactive report and how we can use exactly.

A data model is the physical representation of your tables joined together.
In SQL terms it the FROM and WHERE (JOIN) clauses.
an IR Query section depends on a Data Model. That Data Model can be a Master Data Model (Shared with many Query Sections) or a regular Data Model.
Master Data Models cannot be changed within the Query Section - Their Background is Grey and will indicate so in the upper left corner of the screen...
Locked Data Model: {data model section name}
Master Data Models are useful in that you only design it once and then it and its connection are shared with many Query Sections. And modifications to the Data Model are then applied in the Data Model Section and each dependent Query Section has those changes.
For a more comple defintion open Interactive Reporting and go to HELP. In the Search type "Data Modeling"
You will have plenty to read.
Wayne Van Sluys
TopDown Consulting

ISO: person or forum for data modeling guidance

hello -
Is there anyone here who is willing/able to advise me on the data model that my husband and I have been working on? Alternatively, can anyone suggest an alternative forum, or where I need to go to have this done professionally?
Our background is that we're both self-taught; he has a functioning database (in Approach) that we want to re-build and expand upon. I have a set of related tables in Filemaker that are semi-functional.
We're not sure if his original model is the most stable way to have built it, nor are we sure that we're expanding in the right way to encompass my tables....
I really do NOT want to pay someone to write this whole application for us - beyond the expense, we like/want to be in control.
Thank you -
Marion in Rochester

Marion
My e-mail address is in my profile, drop me a line and I'll be happy to have a look.
A few things first though. Download and install Oracle XE if you have not done so already. Let me know what SQL knowledge you have specifically Oracle DDL (Data Definition Language - Using for creating and altering database objects) and DML (Data Manipulation Language - Used for inserting, updating and deleting rows from a table).
Next send me the data model by way of what tables (and their columns), primary keys, foreign keys, constraints, indexes (if any yet) that you currently have. I will also need to know roughly how many rows you currently have in each table. Send this in any format you want (even just a text file) but not as a database.
Cheers
Ben

DOUBT ABOUT MASTER DATA

Hey guys!!!
I have a doubt about master data, i need to add an atribute to a master data characteristic, do i have to erase the data to do this or can i add the atribute without erase the data?
Thanks and Regards!!

>
Koundinya Karanam wrote:
> Hi,
> You should delete the Master data with SID's first before adding the attribute to that characteristic.
>
> Regards,
> KK.
KK,
Thats not true.
You neednot delete master data to add attributes to a characteristic.
Additionally, you cannot delete (there are methods to delete which are not suggested) master data in normal way, if there is transactional data for that characteristic.
Luis,
You should be able to add attribute to a characteristic without deleting master data, but make sure you load data for the same for values to reflect.

Master Data Mappings

Hi All,
I am trying to map the Datasource to master data InfoObject
The way I did it for 0PLANT_TEXT is:
Right click on the 0PLANT transfer rule and "create transformation"
This pops up the message to either copy Infosource or use the current.
I created a ZPLANT_TEXT infosource and activated that (some routine was a problem while activation and I deactivated the routine)
I then migrated the datasource successfully >created infopackage>created DTP
Is that the right approach?
In BW3.5 this was pretty quick, if I am activating 50 master data objects that would take a lot of time?
I can now see two update rules one is Z update rule for 0PLANT and the other one is the business content. Is it ok to delete to business content one.Why do we need to copy the Infosource??
Thanks,
Gaurav

Is that the right approach? - yeah sure it is
I can now see two update rules one is Z update rule for 0PLANT and the other one is the business content.
Yeah actually there are not two update rules, infact one is update rule and the other is transformation. You can now delete the old update rule(3.x)
Is it ok to delete to business content one.
Yeah it is ok. If you dont want to, you can keep the update rule and never use, but whats the use, go ahead and delete it.
Why do we need to copy the Infosource??
The old is a 3.x infosource and that IS is being used by the object in the update rules. So you would need to create a new infosource. Even when you create a new infosource you have 2 options, 3.x or a 7.0 one.

Master data/text of characteristic 0MATERIAL already deleted

Dear BW Experts:
I'm getting this error message when I tried an extraction into an infocube.
"Master data/text of characteristic 0MATERIAL already deleted"
Before the extraction, I've loaded the characteristic 0MATERIAL succesfully.
Can anyone help me? any clues?
Thanks in advance.

Hi All,
We have also received the "Master Data has been deleted" error message. If you delete (or try to delete) the master data for a characterstic, you will get this error message for the next transaction data load. Just mark the failed request to red and back it out from the data targets and then try loading the data again, this time you will not get any error message.
It will only give that error message for the first load after the deletion of master data.
Now coming to the second point. How to delete the master data?
As all of you know, we delete the master data by selecting delete master data from the context menu of an infoobject. This deletes all the master data records that are not currently being used in any of the infoproviders, hierarchies etc. In the system log (transaction SLG1), it will show what are all the records that got deleted and if a record is not deleted it will let you know where that master data is being used.
Now the problem with deleteing master data in this way is, this runs in dialogue mode and if the no of infoproviders and data volume in those infoproviders is huge it will take very long long time to complete you will get a time out message. There is a function module "RSDMD_DEL_MASTER_DATA" , to run this in background you need to create a ABAP program and migrate it all the way to Production and run it in production.
Luckily, there is a SAP delivered ABAP program which we can use to delete master data. That program is "RSDMD_DEL_BACKGROUND", run this program in back ground with the input parameters as mentioned below.
P_IOBJNM - Infoobject name.
P_CHABNM - Infoobject name.
P_DELSID - X
P_DELTXT - X
P_OBJ - RSDMD
P_SUB - MD_DEL
P_EXT - MDDEL_Infoobject name_BG
You can verify in system log, which records got dleted and which are not deleted.

Difference between master data and transactional data

Hi experts i have just started learning BW can u plz let me know the difference between Master data and Transactional data in bw ? and why is master data is first loaded before the transactional data ? thanks .....

Hi
Transaction data represents any transaction occured between two parties like Customer and Vendor.
We maintain id's to maintain the Customer and Vendor.
When we look at the id we don't have a clue like who is the customer, where does he live,,....
So all this info is maintained in Master data in three categories:
1. Attributes
2. Text
3. Hierarchies.
Collection of Attribute, Texts and Hierarchies is called master Data in BI/BW.
Eg: 0MATERIAL, 0CUSTOMER,,,,,
Master data: is data that remains unchanged over a long period of time. Master data contains information that is always needed in the same way. Characteristics can bear master data in BW. With master data you are dealing with attributes, texts or hierarchies.
http://help.sap.com/erp2005_ehp_03/helpdata/EN/58/a4513ce4af0a40e10000000a114084/frameset.htm
Master Data is one form of the historical data. There is no frequent changes in the data.Every application has it's own Master Data.
In BI the master Data has been divided into
Characterstics
Texts
Hierarchies
We can load Master Data through different ways.
Pls. find the below link
http://help.sap.com/erp2005_ehp_03/helpdata/EN/58/a4513ce4af0a40e10000000a114084/frameset.htm
hope it helps u
regards
gaurav

Modelling 1:N relationships using Master Data

Hello All,
I was trying to store the Bill of Material data in BW and was confused as to how model this as master data. The bill of material is basically a BOM number with a number of components in it. i.e. a 1:N relationship between the BOM and the components.
I was thinking of modelling this by compounding the Component (which has quite a number of attributes) with the BOM.
But practically the scenario is such that, today the BOM contains certain components, but someone can delete a component later in R3. I want this information to be reflected as well in BW, but i suppose BW will not take care of this in the above compounding technique.
What is the best way to model such a scenario in BW. Is there a trick or is the only way to do this is do some checking in the update routines and then delete from the master data tables.
Please let me know your opinions.
Best Regards,
Dabre

Hello All,
Does anyone have a similar kind of experience in modelling a 1:N kind of relationship using master data ? And where the N values could change or be deleted ?
Please share your experiences.
Best Regards,
Dabre

Master data modelling optimum way

Similar Messages

Maybe you are looking for