Post reply

Add column to View or use Named Query in DSV?

  • July 11, 2014 at 3:41 am

    #287805

    I have a Fact table based on a View. Rather than create a Calculated Measure in the front-end BI application (Tableau), I decided to alter the View to include an extra column based on a CASE statement.

    It occurred to me subsequently that I could have left the View untouched and instead replaced the View with a Named Query in the Data Source View.

    My question is: which method is preferred? Does it depend on data volumes?

    I know that you can't use Named Queries in a DSV if the cube storage mode is ROLAP, but that's not an issue because I'm using MOLAP.

    Thanks

    Lempster

  • July 13, 2014 at 10:05 pm

    #1729027

    Lempster (7/11/2014)

    I have a Fact table based on a View. Rather than create a Calculated Measure in the front-end BI application (Tableau), I decided to alter the View to include an extra column based on a CASE statement.

    It occurred to me subsequently that I could have left the View untouched and instead replaced the View with a Named Query in the Data Source View.

    My question is: which method is preferred? Does it depend on data volumes?

    I know that you can't use Named Queries in a DSV if the cube storage mode is ROLAP, but that's not an issue because I'm using MOLAP.

    Thanks

    Lempster

    Personally, I prefer to create the columns in the view. This means that all applications using the view can use the same definition. Otherwise you may need to create an equivalent value in every application that needs it.

    Another reason is that the DSV editor is not real special and will reformat your named query to suit itself. This doesn't help with readability of the query.

  • July 15, 2014 at 7:35 am

    #1729450

    I agree with HappyCat, place it in the view. This gives you (or a DBA) more flexibility in performance tuning the query in the view than a column added in the DSV.

    Thomas

    Thomas LeBlanc, MVP Data Platform Consultant

  • July 15, 2014 at 7:42 am

    #1729451

    Another vote for the view.

    The DSV has the tendency to hide things for you, while a view is more straightforward.

    Also, if the database is in source control it is more obvious to other people that a change took place and that it was the addition of a column.

    Need an answer? No, you need a question
    My blog at https://sqlkover.com.
    MCSE Business Intelligence - Microsoft Data Platform MVP

  • July 16, 2014 at 6:19 am

    #1729812

    Thanks All. So the View it is then. Luckily I defaulted to the 'right' choice! 🙂

    Regards

    Lempster

Viewing 5 posts - 1 through 4 (of 4 total)

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