SQL Server Performance

Avg. Disk Read / Queue Length counter on SQL Server machine jumps to 40-50

Discussion in 'Performance Tuning for DBAs' started by zaurb, Mar 13, 2009.

  1. zaurb New Member

    Hi!

    For couple of weeks we notice that theperformance counter Avg. Disk Read / Queue Length on our SQL Servermachine jumps to 40/50. At this time it's almost impossible to workwith the databases and users have to wait for a few minutes when itdrops to normal 1-2. We have a windows server 2003 with sql server 2000running in the vmware ESX server and SAN.

    I will anyway trydefragmenting from within windows the disk where resides the databasefile but I wanted to know if there could be any other major issues forthis problem. Could this be due to a SAN?

    Thank you!
  2. satya Moderator

    Welcome to the forums.
    I would like to ask on how SAN is configured, is it for WRITES or READS?
    So far I have seen similar issues reporting in particular to the SAN & VMWare environment, whereby the actual process of reporting the average queued I/O read/write statistics is not efficient on the VMware side of it. Since it's not the Windows server itself that's writing the information out to the disk, this is something needs to configured the SAN disks.
    Also I have seen that the most common SAN related performance issues are usually poor selection of RAID level. Raid 0+1 Raid 10 is your best choice when dealing with high I/O operations like databases. The second most common source of issues is not enough spindles (physical disks) to handle the number of I/Os that need processing. You may not need the additional disk capacity but adding additional spindles is a sure fire way to increase read/write performance.

Share This Page