Hitachi Vantara Pentaho Community Forums
Results 1 to 4 of 4

Thread: Why use a Cube? When you have Star Schema's.

  1. #1
    Join Date
    Mar 2006
    Posts
    170

    Question Why use a Cube? When you have Star Schema's.

    Matt and All,

    I'm looking for a business case as to why I would use a cube that is built from the data in the star schemas.
    To explain further I have ETL that sources from OLTP and delivers to an ODS then to Marts ... using Kettle of course :-)

    I am unsure what advantage I would get if any from then proceeding to build or refresh cubes that are sourced from the Marts.

    The only thing I come up with is possibly speed. If the report is sourced from a cube that is FTP’d to them locally then it would be faster then having the report source from the federated db holding the Marts.


    Any input much appreciated.

    Thanks


    Kent Andrews

  2. #2
    Join Date
    May 2006
    Posts
    4,882

    Default

    Reason for cubes... the classic ones as powerplay e.g. I don't think mondrian actually saves the data to an external file.
    - precalculated aggregates/rollups : quicker responses, less use of resources.
    - extra security outside of the database: if you have a portal based solution e.g.
    - historical archiving of cubes
    - ...

    Regards,
    Sven

  3. #3
    Join Date
    Nov 1999
    Posts
    9,729

    Default

    In a lot of ways, especially technically, a cube and a star schema are the same.

    If you however define a cube as a selection from and/or aggregate of a fact table, then there are 2 ways to go:
    1) create aggregate or analytical fact tables and call 'm cubes
    2) use an OLAP engine such as Mondrian (Pentaho Analytics) to speed up queries.

    Personally I think a combination of both techniques is the key to success.
    In a higher sense, what we are doing in a data warehouse, is the enrichment of data.
    We lift data to a higher level. It makes sense to build one fact table/cube on top of another until the data is very refined and easy to consume for higher management. We end up with pre-calculated dashboard information that is easy to consume by higher management, ready to slap on fancy dials and bars. The side effect is that it's easy to consume for OLAP engines as well.

    Allow me to quote the obvious on this subject again:

    Where is the wisdom? Lost in the knowledge.
    Where is the knowledge? Lost in the information.

    -- T.S. Eliot

    Where is the information? Lost in the data.
    Where is the data? Lost in the #@$$%?!& database.

    -- Joe Celko, database consultant/writer
    Remember: it's funny because it's true...

    Matt

  4. #4
    Join Date
    Mar 2006
    Posts
    170

    Smile Sweet! :-)

    Thanks Guys!

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.