DBAsupport.com Forums - Powered by vBulletin
Page 3 of 3 FirstFirst 123
Results 21 to 28 of 28

Thread: Import 300G DB

  1. #21
    Join Date
    May 2001
    Location
    Maryland, USA
    Posts
    409
    Great. So did you use Import for loading data or SQL Loader?
    -- Dilip

  2. #22
    Join Date
    Apr 2001
    Location
    Czechia
    Posts
    712
    Originally posted by Orca777
    how then you will control COMMIT; So you need to define for every Table a PL/SQL-Block.
    Hi,
    I don't think so.
    Direct-load insert doesn't use buffer cache, generates minimal rollback entries and if the target table is nologging, creates just minimal redo for data dictionary changes. Thus, there's no need to commit often, one commit after a table is enough.

    http://otn.oracle.com/docs/products/...1dlins.htm#365


    [Edited by ales on 06-27-2002 at 06:24 AM]
    Ales
    The whole difference between a little boy and an adult man is the price of toys

  3. #23
    Join Date
    Apr 2001
    Location
    Czechia
    Posts
    712
    Originally posted by patel_dil
    Great. So did you use Import for loading data or SQL Loader?
    Are you asking me?
    If yes, I used exp/imp and direct-load insert. No SQL*Loader
    Our databases typically consist of six large tables 1-30 mil rows and tens smaller tables, up to 50,000 rows. I used exp/imp (ROWS=N) to copy structures first, then exp/imp to copy the data of the smaller tables and direct-load insert for the large tables.

    Once I tried to use exp/imp for the large tables and I'll never do it again!
    Ales
    The whole difference between a little boy and an adult man is the price of toys

  4. #24
    Join Date
    Jun 2002
    Posts
    64
    Hi,

    I think export/import is the only way to do cross platform migration.

    Anybody else please...............

    Cheers

  5. #25
    Join Date
    Apr 2001
    Location
    Czechia
    Posts
    712
    With the method above I migrated Sequent->Sun and WinNT->AIX.
    Ales
    The whole difference between a little boy and an adult man is the price of toys

  6. #26
    Join Date
    Nov 2000
    Location
    greenwich.ct.us
    Posts
    9,092
    Several interesting proposals....

    IMHO, since you are switching platforms with a large database, your best bet is to use export/import. You don't really know if your target OS will support files larger than 2G.

    I would all the object owners in one shot, seperating the data into multiple 2000M files. There are several advantages to this method. First, you data will be loaded in the correct order (Which isn't relevent anyway because the constraints are put on at the end). Second, your data will load quicker than across a database link.

    A varient of the following command should suffice:
    exp system/manager file=f1.dmp,f2.dmp,f3.dmp...fN.dmp filesize=2000M owner=schema1,schema2,schema3,...schemaN log=foo.bar

    (I wouldn't use DIRECT=Y in this particular case because you are switching platforms and you have several large export files. There have been some bugs reported with direct=y)

    Remember, when you transfer data from Windoz to Unix, you must ftp in binary mode.
    Jeff Hunter

  7. #27
    Join Date
    Mar 2002
    Posts
    303
    Jeff:
    Is the command:

    exp system/manager file=f1.dmp,f2.dmp,f3.dmp...fN.dmp filesize=2000M owner=schema1,schema2,schema3,...schemaN log=foo.bar

    aqual to >exp .... FULL=Y???

  8. #28
    Join Date
    Nov 2000
    Location
    greenwich.ct.us
    Posts
    9,092
    Originally posted by bensmail
    Jeff:
    Is the command:

    exp system/manager file=f1.dmp,f2.dmp,f3.dmp...fN.dmp filesize=2000M owner=schema1,schema2,schema3,...schemaN log=foo.bar

    aqual to >exp .... FULL=Y???
    I don't like to use FULL=Y with a platform switch.

    Jeff Hunter

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


Click Here to Expand Forum to Full Width