SQL Server Quickie #40 – ColumnStore Indexes

Today I have uploaded SQL Server Quickie #40 to YouTube. This time I’m talking about ColumnStore Indexes in SQL Server.

Like or share to get the source code.

Thanks for your time,

-Klaus

10 thoughts on “SQL Server Quickie #40 – ColumnStore Indexes”

  1. Nice Quickie Sir 🙂

    I have a question which is not related to this but hope you will help me out.

    What is xp_userlock does and its running so frequently on one of my SQL server 2008R2.

    1. Thank You so much for your quick response. 🙂 I already did google but not able to relate it with my scenario. I am getting blocking on my server because of this 🙁

  2. Thanks Klaue.

    Nice and simple presentation and good to know what these different indexes can do. Is there any downside to these sorts of indexes or do they perhaps use more data space ?

    1. Klaus Aschenbrenner

      Hello Andrew,

      Thanks for your comment.
      There are some performance implications that you have to verify, but normally the ColumnStore Index should give you a great performance in reporting workloads.

      -Klaus

  3. Hi Klaus,

    I’me testing the clustered columnstore index right now.
    I’m using the SSDT from VisualStudio 2013.5 for the development.

    When I tried to add a columnstore index:

    CREATE CLUSTERED COLUMNSTORE INDEX ccsidx_yyy ON [dbo].[yyy];
    GO

    I get an error message that:

    Filegroup [XXX] cannot be specified on Table: [dbo].[yyy] when there is the clustered Column Store Index.

    Whats stange that when I ran this command from SSMS the clustered columnstore index has been added to the dbo.yyy table which is using FileGroup.

    Is there any way to solve this issue without moving a table do a different FileGroup (PRIMARY) or to place clustered columnstore index on the whold FileGroup (I’ve found such proposition as well) ?

    Tomasz

  4. Nice and simple presentation and good to know what these different indexes can do. Is there any downside to these sorts of indexes or do they perhaps use more data space ? Can we Create normal clistered index along with column store clustered index?

    1. Klaus Aschenbrenner

      Hello,

      Thanks for your comment.
      No, they use less space because of the applied compression algorithms.
      A traditional Clustered Index doesn’t really make sense.

      Thanks,

      -Klauss

Leave a Reply to Vimal Cancel Reply

Your email address will not be published. Required fields are marked *