Results 1 to 3 of 3
  1. #1
    Join Date
    Jan 2003
    Posts
    4,292
    Provided Answers: 5

    Unanswered: Performance of inplace reorg

    We are using DB2 V8 FP 14 on redhat linux AS 4.0. We have a production server that, I believe needs to have it tables reorged. This server is a 24/7 server so the reorgs need to be "INPLACE ALLOW WRITE ACCESS". I copied the database to a test server through backup/restore. I want to test the impact that the reorgs have before I run them on the production server. I want to check what resources are needed and the speed of the reorgs. I am the only user on the test server and inplace reorg is the only thing running on it. I have 2 very large tables (31 million rows -- 1.65 million pages, 171 million rows -- 3 million pages). The first table took nearly 60 hours to reorg. The second has been going for 4 days and is slightly over 1/3 complete. When I look at the CPU monitor on the server, it never goes over 25%--so the reorg is having very little impact on the server performace. So far, for this last reorg alone, it has generated almost 75 GB of log files--this I expected.

    My question is, can I do something to get the reorg to go faster?

    Thanks,

    Andy

  2. #2
    Join Date
    Mar 2007
    Posts
    35
    Quote Originally Posted by ARWinner
    We are using DB2 V8 FP 14 on redhat linux AS 4.0. We have a production server that, I believe needs to have it tables reorged. This server is a 24/7 server so the reorgs need to be "INPLACE ALLOW WRITE ACCESS". I copied the database to a test server through backup/restore. I want to test the impact that the reorgs have before I run them on the production server. I want to check what resources are needed and the speed of the reorgs. I am the only user on the test server and inplace reorg is the only thing running on it. I have 2 very large tables (31 million rows -- 1.65 million pages, 171 million rows -- 3 million pages). The first table took nearly 60 hours to reorg. The second has been going for 4 days and is slightly over 1/3 complete. When I look at the CPU monitor on the server, it never goes over 25%--so the reorg is having very little impact on the server performace. So far, for this last reorg alone, it has generated almost 75 GB of log files--this I expected.

    My question is, can I do something to get the reorg to go faster?

    Thanks,

    Andy

    turn on intra parallel and increase utilitiy_heapsz as much as possible for example 100000,
    please let me know about the results
    good luck,

  3. #3
    Join Date
    Jan 2003
    Posts
    4,292
    Provided Answers: 5
    Intra_parallel is already on. From the documentation for UTIL_HEAP_SZ:

    This parameter indicates the maximum amount of memory that can be used simultaneously by the BACKUP, RESTORE, and LOAD (including load recovery) utilities.

    Reorg does not use this, so increasing it will naot have any affect.

    Andy

Posting Permissions

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