Tape Backup or Disk Backup Solution

Discussion in 'HP' started by ZooOYork, Mar 13, 2006.

  1. ZooOYork

    ZooOYork Guest

    Hi,

    I am not sure if posting in the correct group but here is my problem.
    We currently have 3 tape backups connected to a server.
    I am having a hard time getting all the jobs to run in time and
    juggling tapes because all the data does not fit on 1 tape.
    all 3 drives are 200/400 hp ultrium drives. We do FULL backups
    no incrementals or differentials. we do not want to deal with multiple
    tapes. I have not worked with any autoloaders as i heard they only can
    backup
    1 job at a time. what would be a good solution? backup to disk first?
    then backup to tape?

    any ideas?

    thanks in advance
     
    ZooOYork, Mar 13, 2006
    #1
    1. Advertisements

  2. ZooOYork

    Hoff Hoffman Guest

    :We currently have 3 tape backups connected to a server.
    :I am having a hard time getting all the jobs to run in time and
    :juggling tapes because all the data does not fit on 1 tape.
    :all 3 drives are 200/400 hp ultrium drives. We do FULL backups
    :no incrementals or differentials. we do not want to deal with multiple
    :tapes. I have not worked with any autoloaders as i heard they only can
    :backup 1 job at a time. what would be a good solution? backup to disk
    :first? then backup to tape?

    You can upgrade the storage capacity of the tape drive (if feasible), or
    you can decrease the volume of data (incrementals, etc), or you can move
    to a media loader or a library if your particular software environment
    supports continuation onto additional storage media.

    Disk-based operations can provide a way to use a disk as archival media
    or as a way to cache the archival data on the way out to other media;
    as an archive, or as a cache to reduce the size of the archival window.
    (Though I will mention that many current-generation tapes have higher
    available bandwidth than many current-generation disks, and many archival
    operations are accordingly limited by the I/O throughput -- file system
    or database access, and/or fragmentation and/or I/O contention, etc --
    on the input, so this approach doesn't always work to your advantage.)

    Loaders and Libraries are limited by the numbers of drives contained
    within, and ancillary concerns around software support, etc. Various
    of the low-end devices have a single drive, and a way to sequentially
    (and sometimes randomly, if and as requested) load cartridges.

    If you have a sufficient archival window for your current processing,
    then a loader or a library can work. If not, you need to go parallel
    and/or to less data written and/or to faster I/O device(s) and/or to
    better I/O bandwidth between the source and the sink, err, the archival
    device.

    There are also network archival operations, but the considerations
    are basically the same -- though with the addition of client software,
    a network pipe, and software on the archival server, obviously.

    ---------------------------- #include <rtfaq.h> -----------------------------
    For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq
    --------------------------- pure personal opinion ---------------------------
    Hoff (Stephen) Hoffman OpenVMS Engineering hoff[\0100]hp.com
     
    Hoff Hoffman, Mar 13, 2006
    #2
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.