Re: LMT Autoallocate initial, tot # of extents

  • From: Barbara Baker <barb.baker@xxxxxxxxx>
  • To: Niall Litchfield <niall.litchfield@xxxxxxxxx>
  • Date: Wed, 8 Jun 2005 08:02:14 -0600

Niall:
When I said I agonized over autoallocate vs uniform, I really meant
agnoized.  I waffled back-and-forth for several weeks, reviewing the
discussions on this list, and essentially not getting anywhere.  I
have thousands of objects to move from DMT to LMT.  In the final
analysis (and I know this is a severe cop-out), I decided that I just
don't need the grief!  The application that sits on top of this
database is a heap o' crap, so it just didn't seen to matter that
much.  Nearly impossible to tell how objects are going to grow/shrink.
  So I decided to quit waffling and just do something.


For the record, I continued experimenting with this table.
When I asked for an initial of 256m, I got an initial of 8192k, total
198 extents

When I asked for an initial of 2048m, I got an initial 64m, total 142 exten=
ts

At least I'm not still stuck with an initial of 1024k  ;-))

On 6/7/05, Niall Litchfield <niall.litchfield@xxxxxxxxx> wrote:
> Hi Barb
>=20
> On 6/6/05, Barbara Baker <barb.baker@xxxxxxxxx> wrote:
> > Oracle 9.2.0.4        Solaris 9     Blocksize 8192
> >=20
> > I'm attempting to move (alter table .. move tablespace)  several=20
> > tables from DMT to autoallocate LMT for a database migrated from
> > 8.1.7.4 to 9.2.0.4.  (Agonized over autoallocate vs. uniform size.=3D20
> > Perhaps I made the wrong choice.)  Anyhow . . .=20
> >=20
> > I have 1 table in particular that's 2752 megs 185 extents in the DMT.=
=3D20
> > After I move it to the LMT, it's 2757M,  211 extents.  Why so many
> > extents??    I'm not as concerned with the # of extents as I am in
> > understanding what it's doing.
> =20
> If you choose autoallocate over lmt you choose to give up understanding o=
f
> what is happening in favour of easier admin. You got 211, someone else mi=
ght
> get 200 and 2800mb and so on. You don't know. if you care or not is up to
> you. Not caring is valid, but not a choice I make.=20
> > --=20
> > Niall Litchfield
> > Oracle DBA
> > http://www.niall.litchfield.dial.pipex.com=20
>
--
//www.freelists.org/webpage/oracle-l

Other related posts:

  1. Home
  2. oracle-l
  3. Archive
  4. 06-2005