Skip to Content
author's profile photo Former Member
Former Member

Master data performance

Hi Gurus,

I have some concerns with Data Modeling a master data Infoobject into an Infocube. The master data object will have hundreds of millions of rows. It is for my realignment process that will have ship-to/material, so I really have no choice about the number of records I need to store.

Does anybody have any data that can support the general experience in terms of performance of :

a) adding records to the MD object and

b) querying the Cube?

Shall I store the MD object as a line-item dimension for better query performance?

Regards,

Alex

Add a comment
10|10000 characters needed characters exceeded

Related questions

4 Answers

  • Best Answer
    Posted on Oct 20, 2006 at 07:10 AM

    Hi,

    I do not understand really well your problem...

    You have got millions of rows in a MD like MATERIAL with a compound SHIP TO.

    So you should avoid having both of them in the same dimension of your cube if the same MATERIAL code can be delivered to several SHIP TO partner. Otherwise, no problem to put them together into the same dimension.

    regards

    Add a comment
    10|10000 characters needed characters exceeded

  • author's profile photo Former Member
    Former Member
    Posted on Oct 20, 2006 at 08:08 AM

    Your re-alignment process will be severely impacted if you have aggregates that use the infoobject.

    Can you explain a bit more about what this "object" looks like...and how you are modeling it?

    Cheers!

    /smw

    Add a comment
    10|10000 characters needed characters exceeded

    • Former Member Laurent THIBERT

      Hi Laurent,

      The issue is with how many records I fear to add to this compound infoobject. It will be a line-item dimension, however its size could be as many rows as what is on the fact table.

      My concern is with regards to how the overall performance of inserting records to the Master Data object will be. What about querying? Are there other Data models that you can recommend?

      Regards,

      Alex

  • Posted on Oct 23, 2006 at 06:49 PM

    Hi Alex:

    In addition make that dimension as "High Cardinality" in addition to Line-Item dim.

    Laoding Master Data has nothing to do with many records in CUBE. So, you cube load will be slower.

    You should look into CUBE partitioning, etc to fine tune queries on CUBE.

    Ram C.

    Add a comment
    10|10000 characters needed characters exceeded

  • author's profile photo Former Member
    Former Member
    Posted on Oct 23, 2006 at 07:25 PM

    Is there any way you can break out the data into mulitple infocubes, that all look the same, except for the data?

    For instance, If you have multiple sales orgs, you could create mulitple infocubes (one for each sales org), and put a multiprovider on top of it, to keep the info contents low, but the data will only be accessed when needed (it wouldn't all be stored in the same tables....).

    That may help performance.

    /smw

    Add a comment
    10|10000 characters needed characters exceeded

    • Former Member

      Thank you both, points assigned.

      Steve, I actually did propose this as an alternate solution to my client. I also proposed a Cube that contained the Master data object's attributes, that could be MC'd with the Sales cubes. Another solution proposed was to house the data in an ODS, that could be Joined (left-outer join) with a InfoSet.

      Regards,

      Alex

Before answering

You should only submit an answer when you are proposing a solution to the poster's problem. If you want the poster to clarify the question or provide more information, please leave a comment instead, requesting additional details. When answering, please include specifics, such as step-by-step instructions, context for the solution, and links to useful resources. Also, please make sure that you answer complies with our Rules of Engagement.
You must be Logged in to submit an answer.

Up to 10 attachments (including images) can be used with a maximum of 1.0 MB each and 10.5 MB total.