Post reply

Snowflake dimension design question.

  • February 9, 2009 at 6:52 am

    #383896

    I am trying to get an undertstanding of Snowflake dimensions and how they could benefit the design of a DW.

    Snowflake dimensions appear to me to be used to normalise dimension tables and therefore removing repetition of data in Dimension records.

    I believe the Dimension -> Snowflake is a 1 to 1 relationship. Is this correct?

    Is it advisable to use Type 2 changing dimension techniques on changing the Snowflake (I assume this requires the Dimension to also be Type 2) to track changes or to kepp them always to Type 1?

    Thanks

    Daniel

  • February 10, 2009 at 11:39 am

    #941508

    When I design data warehouse, I try to avoid snowflake dimension because it requires an extra join to get the data. Sometimes it is better to de-normalize the database to get a better result. Ralph Kimball talked about this in his book and he also recommended not using snowflake.

Viewing 2 posts - 1 through 1 (of 1 total)

You must be logged in to reply to this topic. Login to reply