Re: So exactly what IS clustering?

From: Tom Moreau (tom_at_dont.spam.me.cips.ca)
Date: 04/11/04

  • Next message: Rob Oldfield: "Re: So exactly what IS clustering?"
    Date: Sun, 11 Apr 2004 19:00:22 -0400
    
    

    I'd go with .server and .programming for this type of question.

    As for your app, the rule is - test, test, test. Generally, adding a few
    indexes doesn't hurt. Adding a lot does. Also, your app may not
    necessarily use the indexed views unless you are using the Enterprise
    Edition of SQL Server.

    Let us know if the new indexes made a difference. Also, if you are using
    clustered primary keys, the ITW isn't smart enough to consider changing a PK
    from clustered to nonclustered and then clustering on something else. You
    may want to experiment there.

    -- 
       Tom
    ----------------------------------------------------
    Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
    SQL Server MVP
    Columnist, SQL Server Professional
    Toronto, ON   Canada
    www.pinnaclepublishing.com/sql
    .
    "Rob Oldfield" <rob@oldfield100_wow_freeserve_yikes_co_incredible!_uk> wrote
    in message news:OOR24eBIEHA.2376@TK2MSFTNGP12.phx.gbl...
    Thanks for the info.  I've already saved the recommended script but haven't
    run it yet.  In practical terms my problem is really whether I need to
    change anything in my app in order to take advantage of the new indexes.
    (And given that the ITW is reporting that it will give a 34% performance
    increase it has to worth looking at).  Not being a major SQL server expert I
    can't tell which newsgroup is best for telling the difference between
    different types of view.  Any ideas on that one?
    Thanks again.
      "Tom Moreau" <tom@dont.spam.me.cips.ca> wrote in message
    news:Ot49L8AIEHA.3476@TK2MSFTNGP11.phx.gbl...
      There's clustering and then there's clustering.  ;-)
      The clustering to which this ng refers is server clustering, whereby one
    server acts as the backup for another.  The backup automatically picks up
    when the primary server fails.
      The clustering to which you refer is index clustering.  This is basically
    the physical sort order of your table.  Thus, there can be - at most - only
    one clustered index per table.  Also, in SQL Server 2000, you can have
    indexed views.  Here, you create a unique clustered index on the view, which
    materializes the view.  You can also create nonclustered indexes on:
      all tables
      indexed views which already have their clustered index built.
      The ITW will script the recommended indexes for you.  You should save
    these to a file which you can then run at any time.
      -- 
         Tom
      ----------------------------------------------------
      Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
      SQL Server MVP
      Columnist, SQL Server Professional
      Toronto, ON   Canada
      www.pinnaclepublishing.com/sql
      .
      "Rob Oldfield" <rob@oldfield100_wow_freeserve_yikes_co_incredible!_uk>
    wrote in message news:%237$nwAAIEHA.2260@TK2MSFTNGP09.phx.gbl...
      Hi all,
      Some very basic questions here.  I have a VB.Net app that is using an SQL
    2K
      server as it's data source.  Everything is working fine but it's a little
      slower than I would like.  So I've just run the index tuning wizard over
    the
      SQL instance that I'm using.
      The result of that is a list of my tables specifying my existing indices
      (it's pretty straightforward stuff, so just the existing PKs), each of
    which
      has a tick in the 'clustered' column.
      In addition, it's suggesting two new indexed views, one of which has a
    tick
      in the 'clustered' column, the other doesn't.
      My questions:
      What is clustering?  I've looked through books on-line but can't find
      anything that explains it to me.  Can somebody explain it or have a
    resource
      that does?
      Are my existing tables and views clustered?  Or is the index tuning wizard
      suggesting that I switch them (the tables) to be clustered?  I've just set
      them up in EM without doing anything clever.
      And one last question that doesn't really fit this newsgroup, but
    hopefully,
      isn't too off topic: so far I haven't told the index tuning wizard to
      actually put its recommendations in place - will that just create the
      relevant views; is there a way of seeing what those views look like before
      taking this step; and, in general terms, is applying the results of the
      tuning wizard a safe activity, or do I really need to back everything up
      first?
      Thanks for any help.
    

  • Next message: Rob Oldfield: "Re: So exactly what IS clustering?"

    Relevant Pages

    • Re: Newb Active-Passive / Active-Active Single Instance Question
      ... The Disk space is SAN storage. ... node and one virtual server. ... cluster node as it is no longer like Windows Clustering is it that SQL 2005 ... Also, SQL Server ...
      (microsoft.public.sqlserver.clustering)
    • Re: Using sql server clustering
      ... Clustering will give that one high availability. ... other 4 can still handle the load while you fix the one that is down. ... add more server for incoming data wich is failover cluster. ... About the servers to pull data, can i use sql server clustering in order ...
      (microsoft.public.sqlserver.clustering)
    • Re: Memory Allocation - Win2k3 SP1 and SQL Server 2K SP4
      ... Microsoft SQL Server MVP ... MVP - Windows Server - Clustering ...
      (microsoft.public.sqlserver.clustering)
    • Re: So exactly what IS clustering?
      ... I'd go with .server and .programming for this type of question. ... Columnist, SQL Server Professional ... So I've just run the index tuning wizard over the ... What is clustering? ...
      (microsoft.public.sqlserver.clustering)
    • HA & Failover options
      ... instances for all the technology components (App Server, DB Server, ... Application Server 10g Release 3 and Oracle Database Release 2. ... Clustering features. ... how instantaneous is failover to the secondary site with data ...
      (comp.databases.oracle.server)