Post reply

Trans Log Backups killing the server

  • June 9, 2005 at 3:21 pm

    #564411

    Update:
    Running the transaction log backup job to write to the shared drive on the dataclstr  improved performance at noon this afternoon while the job was running.  Of course there was still high cpu utilization but the system was much more responsive to user requests and the job did not take any extra time to run.
     
    I wouldn't say this solves the performance problem completely but obviously performance gains were made when we dealt with the disk contention possibility.  Our next steps are:
    1) to add a cpu/ram expansion board to the servers in the cluster.
    2) the developer is making a change to their code that will cut down on a disk read action in their code that is mostly not needed.
     
    Thanks everyone for your previous input!  Mucho Gracious!
     
     
    Paula
     
    PS.  Rick!  Microsoft told me on the phone today that they released a fix last night for the sp4 AWE bug problem - just wanted to let you know.

Viewing post 16 (of 15 total)

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