SQL Server Performance

SQL 2008 clustering

Discussion in 'SQL Server 2008 Questions [Archive Only - Closed]' started by WingSzeto, Sep 30, 2008.

  1. WingSzeto Member

    Does SQL 2008 clustering as far as functionality remain the same as SQL 2005? Do they still have the same 'best practises' for the quorum disk? I have three set of physical disks. One is for the database (raid 10), one is for the Tempdb (raid 1), and one is for the tran log (raid 1). Unfortunately, we don't have a separate set of physical disks for the quorum. Our database size is about 45 GB and under performance counter SQLServer:Databases - transaction per sec for our production database is 24/sec. The same counter for tempdb is about 235/sec. The read and write queue length counter for the log file drive is very low, < 0.01.
    Based on our disk setup, which set of disks is more preferable to be used for creating a logical disk for the quorum? Also, is there some best practise whitepaper for setting up SQL 2008 clustering?
  2. MohammedU New Member

    It is the same as SQL 2005 but quorum concept is different in windows 2008.
    I will go with raid 10 as quorum drive but Quorum model has be determined based on your HW...
    Windows Server 2008 Failover Clustering - The New Quorum Model

    One of the big changes in Windows Server 2008 Failover Clustering is the new quorum model. In Windows Server 2003, we had only two choices, either the single disk quorum that has been around since NT 4.0 or Majority Node Set (MNS). Actually, there are three if you consider MNS with the File Share Witness (FSW) as a separate option.
    In Windows Server 2008 Failover Clustering, administrators now have four choices on how to implement the quorum.
    • One option is to use Node majority. In this option, a vote is given to each node of the cluster and the cluster will continue to run so long as there are a majority of nodes up and running.
      • A second option is to use both the nodes and the standard quorum disk. In this option, a common option for two node clusters, each node gets a vote and the quorum, now called a witness disk) also gets a vote. So long as two of the three are running, the cluster will continue. In this situation, the cluster can actually lost the witness disk and still run.
        • A third option is to use the classic/legacy model and assign a vote to the witness disk only. This type of quroum equates to the well known, tried, and true model that has been used for years.
          • A fourth option, is, of course, to use the MNS model with a file share witness.
        • It has been a few days since I have seen the GUI, so I can't tell you off the top of my head which order they appear in within the GUI.
          Two notes that caught my attention the other day when talking about these options is that it is not possible to use DFS as the file share witness and with changes to the quorum there aren't any checkpoints so there is no longer a need for the -resetquorumlog switch on starting the cluster service.
  3. WingSzeto Member

    Thank for the information. I like to ask you DTC set up in SQL 2008 also. Again, I don't have any more physical disks available for DTC.
    Is it ok to share DTC with one of the SQL disks? During the DTC configuration, it asks for a disk storage to be put on.
    Another issue I am seeing for sharing disk between DTC and a SQL data disk is that when a disk is allocated to DTC, SQL server won't see the disk until I move that disk used by DTC back to the SQL application resource group. But as a result, DTC will be part of that SQL resource group. Some whitepaper suggested that DTC should be in its own resource group, which so far I couldn't find a way to make it happen. Is it ok to have DTC in a SQL resource group not in its own group?

Share This Page