Re: Algorithm for calculating extent size in LMT

Connor,

That's a slightly misleading, or perhaps I should say over-optimistic, test case
because all the extents are probably
sequentially adjacent. I set my test up to avoid the accidental benefit of Oracle having a concealed optimisation for the
special case of consecutively numbered
extents being physically adjacent in the bitmap - maybe Oracle is dropping 8
extents at a time in your example, and
reducing its logging requirements as
a consequence.

Remember the primary rule of testing
Oracle:

    "Why be awkward, when with a little
    care and attention you can be an
    absolute bastard."

Mind you, as you pointed out for someone else's benefit on Metalink recently - I bet you didn't have a tablespace quota when
you did that one !

--
Jonathan Lewis
http://www.jlcomp.demon.co.uk

Now running 3-day intensive seminars
http://www.jlcomp.demon.co.uk/seminar.html

Host to The Co-Operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html

Author of:
Practical Oracle 8i: Building Efficient Databases


Connor McDonald wrote in message <3C851D76.21EC_at_yahoo.com>...


>Similarly

>

>SQL> create table test_table ( x number )

>  2  storage ( minextents 50000);

>Table created.

>SQL> set timing on

>SQL> drop table test_table;

>Elapsed: 00:00:05.09

>

>LMT's.... Love 'em!