lkml.org 
[lkml]   [1998]   [Jun]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: (reiserfs) Re: LVM / Filesystems / High availability
    Hi,

    In article <358E5A5D.CA854A04@vadim.gem.net>, "Vadim E. Kogan"
    <vadim@vadim.gem.net> writes:

    > I see some problems in current setup. The problem is that raid0/1/5
    > provide protection from disk failures, not from crash or unclean
    > shutdown. With Journaling FS, that knows how to use multiple devices,
    > it's possible to archieve all 3 - speed, protection from disk failures
    > and protection from crash/unclead shutdown. As you understand, all 3
    > are important.

    Yes. 2.3 will definitely get ext2 journaling.

    >> Absolutely. In fact, I'm beginning to think that linux-2.4 should
    >> actually be numbered 3.0, with "Enterprise Scalability" being the big
    >> thing for the new version number. The main thing needed for that over
    >> and above what's already been described is more scalable SMP.

    > Linux will have to include many more functions before it'll be usefull
    > for "Enterprise" level. Some of these functions are already developed
    > and live separately, some are being developed now and aren't in usable
    > stage yet, some are not even started.

    What functions do you think are missing? If you compare the Linux
    kernel right now with what is in NT, then all of the things which NT
    calls Enterprise Level are mostly already in Linux, and 2.4 will almost
    certainly be a match for NT on all accounts. SMP performance (more fine
    grained locking in several places), filesystem performance
    (journalling/btrees) and overcoming the current device restrictions for
    things like SCSI and ttys are all expected for 2.3.

    The fact is that a lot of people think of NT in terms of Enterprise
    level functionality, and if we have the same level of performance and
    functionality on Linux then we'd be crazy not to let people know about
    it. The Linux kernel is really maturing quite remarkably; most of the
    interesting problems left for Linux in the enterprise and on the desktop
    are in user space. The only really big thing left is likely to be HA
    clustering, as in MS's Wolfpack, but then NT server doesn't give you
    that by default anyway.

    --Stephen

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.rutgers.edu

    \
     
     \ /
      Last update: 2005-03-22 13:43    [W:0.021 / U:89.156 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site