Post reply

'Lock pages in Memory' on 64 bit w failover active/active clustering

  • May 14, 2010 at 7:28 am

    #221099

    My questions in regard active/active failover clusters we are moving to a 64 bit machine and how the locked memory work when a cluster fails.

    Examples: If the first active cluster is locking 60GB in memory and then fails, does the new active cluster pull an additional 60GB due to the Lock Pages In MemoryIn Memory, meaning 120GB needs to be available, or will it drop the first locked 60GB and then relock it for this new cluster?

    Thanks!

  • May 14, 2010 at 8:27 am

    #1167572

    An Active/Active Cluster means that you have two sets of clustered instances, one running on the first server, the other on the second.

    When one server/instance fails, that instance starts on the other server, so two instances are running on a single server, so there needs to be enough resources to run both concurrently from a single node or you're entirely negating the redundancy of the solution.

    The two sets of instances are entirely independent of each other and one does not release memory to accommodate another in the event of a failover.

    The way you phrase the question suggests that you need to do some reading on the general architecture of Active/Active Failover clustering before going much further with this!

  • May 14, 2010 at 8:30 am

    #1167578

    Thanks Howard, I understand you answer. This was the answer I expected. You are correct I need more knowledge, I am just a DBA that was given the duty to reasearch this topic by my Data Architect. 🙂

  • May 14, 2010 at 8:35 am

    #1167581

    No problem. Glad I could help 🙂

Viewing 4 posts - 1 through 3 (of 3 total)

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