Post reply

SQL Instances Performance

  • June 1, 2005 at 9:10 am

    #88521

    All,

    Our SQL box has 7 instances of SQL Server running on it.  One particular instance is a lot slower than the rest.  All instances have the same database's although the sizes differ.

    I had been running performance checks against the DB, in particular, every index I have checked has a fill factor of 0.  I have been looking at various other area's.  I ran a DBCC CHECKDB and nothing comes up.

    I ran a DBCC SHOWCONTIG on a table with 65k rows and it took 8 minutes to process.  I ran the same command on the same table, but in another instance on the same server database.  This table had 45k rows and it took 3 seconds!  I'm a bit alarmed by the difference in how long it takes to process this command.

    I'm a bit stuck now with where I should be looking.  Could this be physical disk fragmentation?  All of the databases are in a SAN.

    Any other suggestions?

    Thanks,

    Clive

  • June 1, 2005 at 9:30 am

    #562229

    First thing that comes to mind is how much RAM is associated to each instance.  With 7 instances the minimum RAM setting may just be leaving a couple of the starved.

  • June 1, 2005 at 9:41 am

    #562236

    The server has 8Gb of memory and each instance is configured to dynamically allocate memory.  There is on limit on how much each instance can use.

    This is the config of company who designed the whole system.  It works this way for all of the other SQL boxes and there are no problems of this nature on them.

  • June 2, 2005 at 3:27 am

    #562452

    Just an update...

    Done a complete site DBREINDEX and updated all of the statistics.  Optimizations also ran for the DB's on the instance.

    Although the connectivity is now better, it is still slow compared to what it should be.

    Noticed that navigating through EM is still very slow.  For example, editing a scheduled job, takes approximately 1m30s to open up after double clicking on it...!

    Considering a move to a new SQL Server now

  • June 3, 2005 at 5:26 am

    #562822

    Another update...

    It appears that the cluster node that the database was on was overloaded.  We failed the resourse over to the other node and everything was back to normal.

    Think it's time I started to learn about clustering!!

    Clive

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

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