Slashdot Mirror

← Back to Stories (view on slashdot.org)

GFS, OCFS, and GPFS - Which Filesystem for Oracle?

Posted by ryuzaki0 on from the proper-care-and-feeding-of-a-high-octane-RDBMS dept.

amani asks: "My company has a Oracle 9i RAC database running on a Sun cluster. In 6 months we are looking to replace the cluster with either a Linux or an AIX solution that will involve SAN storage. I see that their are a variety of filesystems for Oracle and Linux. Sistina (Red Hat) has the GFS, Oracle has the OCFS, and IBM has GPFS. Does anyone know the pros and cons of each of these filesystems ,and which one would be better for a continuously growing database?"

3 of 36 comments (clear)

  1. Re:choose AIX/JFS/SSA by Anonymous Coward · · Score: 1, Interesting
    100 Gig?

    Simple SW raid with 3X 300GB IDE drives for growth
    ext3, metadata journaling only,
    and I'd toss oracle and use postgresql.

    3TB is more interesting.

    (more realistically, yes, I have a 100G postgresql, ext3 database that works fine. I also have a 5GB oracle database on some veritas file system. What does the size of 100 Gig matter?)

  2. Parting with Sentiment by fm6 · · Score: 3, Interesting
    If you have lots more updates than accesses, you need your redo logs etc on RAW devices, no filesystem required, these will be your biggest bottleneck.
    OK, but that sort of begs the question. One of the filesystems mentioned OCFS, is specifically designed to use in place of a raw partition. So when is a raw partition preferred and when OCFS?

    Despite all the wisecracks about the name, our sentimental favorite should be GPFS because of a certain well known geek who works for the filesystem group at IBM Almaden.

  3. Re:General Protection FaultS? by Anonymous Coward · · Score: 1, Interesting
    I'd avoid it for another reason. I used to work somewhere it was implemented in an NFS cluster using HACMP. It sucked. Royally. I managed to crash the cluster simply by doing some rsyncs in parallel; the processes crashed one node, when it failed over it promptly crashed the other node bringing an entire 10TB cluster down.

    Last I heard, they had a short time to fix it before the lawyers got involved, but I think IBM had pretty much given up and were looking to pay for a NetApp replacement just to keep them happy.

    I will say, however, that I don't know how it works with Oracle.

    Posted anonymously to protect the inno^H^H^H^Hguilty.