DBAsupport.com Forums - Powered by vBulletin
Results 1 to 9 of 9

Thread: PCTINCREASE ADVISE

  1. #1
    Join Date
    Aug 2000
    Posts
    132
    Hi all,

    I know that the DBA handbook advises that a zero PCTINCREASE on tablespaces. Our System, RBS and Data tablespaces have a PCTINCREASE of 50 - are these the defaults. Should I be concerned? Should/Can I change them?
    I believe our consultant set them up that way but I'm not sure why.

  2. #2
    Join Date
    Apr 2000
    Location
    Baltimore, MD
    Posts
    759
    I believe the default is 50. You need not be concerned, and yes, you can change them.

  3. #3
    Join Date
    Jun 2000
    Location
    Madrid, Spain
    Posts
    7,447
    Zero is better since you get uniform extent size, uniniform extent size leads to freespace fragmentation

  4. #4
    Join Date
    Aug 2000
    Posts
    132
    could it be that our consultant set up the system, rbs, temp and data tablespaces with the default b/c he wanted to allow those tablespaces to extend? Is it recommended to change it to 0?

  5. #5
    Join Date
    Nov 2000
    Location
    greenwich.ct.us
    Posts
    9,092
    PCTINCREASE doesn't really have anything to do with tablespaces extending.

    PCTINCREASE deals with the segments inside the tablespace. For example, if you have a table that has a next extent of 10K and PCTINCREASE is set to 50, the next-next extent would be 15K. The next-next-next extent would be 22.5K, etc.

    If your PCTINCREASE is set to 0, all extents would be 10K, thus uniform extent sizes.

    I like to set the default PCTINCREASE at 1 for the tablespace and then all the objects have a pctincrease of 0. This way, I can tell when my developers sneak a table in without sizing it correctly and don't have to worry about the table extending out really quickly.
    Jeff Hunter

  6. #6
    Join Date
    Sep 2000
    Posts
    384
    A non zero value will ensure that tablespace coalesce happens as and when smon wakes up.But the problem with 0 is that you have to manully coalesce the tablespace.
    for temp and rbs keep pctincrease as zero .for all others make it 1.
    Radhakrishnan.M

  7. #7
    Join Date
    Jun 2000
    Posts
    417
    is is better in the long run to have a script to coalesce your tablespaces instead of having to deal with the non-uniform extent sizes from the pctincrease?

  8. #8
    Join Date
    Jun 2000
    Location
    Madrid, Spain
    Posts
    7,447
    If you have uniform extents you will never have to coalesce that is why when pctincrease is zero SMON doesnt coalesce the extents because simply there is no need to.

  9. #9
    Join Date
    Sep 2000
    Posts
    384
    You create the segments with pctincrease as zero and for tablespace 1.
    Radhakrishnan.M

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