Post reply

I have quick question regarding Cube implementation on production

  • January 2, 2008 at 5:22 pm

    #181723

    I have quick question regarding Cube implementation on production

    Environment

    We have two production servers

    1 BCP servers – Backup every weekend form Production server

    2 Production servers (live Data)

    Cube Environment

    Analysis Services - SQL 2000

    Approximate size of the cubes - 10GB - 2 Cubes, 2GB - 2 cubes

    Full Process Time 6 Hrs

    Incremental Update time ½ Hrs

    I have modified the existing cubes to incorporate the new dimension structure

    I am planning to do the following steps to implement the cubes

    Step 1 Modify the cube on Development server

    Step 2 Unit test on testing servers

    Step 3 Backup the cube and restore the cube on BCP Server

    Step 4 Backup the cube from BCP Server and restore the cube on Production Server and re-process the cubes (Full Process)

    Or

    Step 4 Backup the cube from BCP Server and restore the cube on Production Server

    And do the Incremental Update

    I believe full process is the best method but it takes 6 hrs to process the cube

    Could you please suggest the best method?

    Best regards

    Vijay

  • January 2, 2008 at 11:51 pm

    #764266

    Are u adding a new dimension or an attribute to an exisiting dimension.

    if ur adding a new dimension u need to go with process full.

    As a thumb rule i would suggest when making changes to the cube design always run Porcess full once , when making chnages to only underlying data use incremental process.

    Hope this helps

    J

    Jayanth Kurup[/url]

  • January 3, 2008 at 3:06 pm

    #764590

    Hi Jaykay

    Thanks for the reply. Yes I am adding 5 new virtual dimensions.

    You are absolutely correct. I have to run full process

    Best regards

    Vijay

Viewing 3 posts - 1 through 2 (of 2 total)

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