Hi,
  This is from a metalink note 1039341.6. It is about 'DEFAULT STORAGE'. I do not know if the OCP question was regarding that.
***********

Use the following guidelines to specify DEFAULT STORAGE:     Set INITIAL=NEXT.Since a process always writes data equal to  SORT_AREA_SIZE to a temporary segment, a good value for the extent size is (n*s + b)

    where:  n is a positive integer 

            s is the value of SORT_AREA_SIZE initialization parameter

            b is the value of DB_BLOCK_SIZE initialization parameter

    Using this value optimizes temporary segment usage by allowing sufficient space for a header block and multiple sort run data to be stored in each extent.

************
Is this not correct? Forget OCP. I was just want clarify this issue.  Good luck for the exams  Prem.

Further to that if I give my sort area size as a very odd number which is not a multiple of OS block size then what happens.

Thank You
[EMAIL PROTECTED] wrote:

Prem ,
OCP has nothing to do with knowledge and nothing to do with the real world.
So if you want to give OCP, forget what you know and humbly, play by its
rules.

Right now if your are serious about giving OCP exams then please
enable "suspension of disbelief" , don't ask why and just mug up the
following 3 (any one of them may come up in the ocp exam , though not
together) :

1.
To optimize the performance of a sort in a temprary tablespace , set the
UNIFORM SIZE to be a multiple of the parameter sort_area_size

2.
To optimize the performance of a sort in a temprary tablespace , set the
UNIFORM SIZE to be a multiple of sort_area_size * db_block_size

3.
To optimize the performance of a sort in a temprary tablespace , set the
UNIFORM SIZE to be n * sort_area_size + db_block_size.


OCP has nothing to do with knowledge and nothing to do with the real world.
So if you want to give OCP, forget what you know and play by its rules.
Best of luck .




Do you Yahoo!?
Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes

Reply via email to