Re: table with keep as buffer pool see much more physical reads than the number of blocks in the table

I think Mark's comments are probably in the right area. If you update the table in one session, then other sessions are going to start creating read-consistent clones. Do you have a recycle pool ? If not, then the clones will end up in the Keep pool.

It's possible that if you generate too many clones of blocks 1 to 100 (say) then the "originals" of some other blocks have to be kicked out of memory to make way for them - and then get read back later.

You could check with a query like:

select file#, block#, count(*)
from v$bh
where objd = {data object id of table} group by

         file#, block#
having

        count(*) > 1
/

to see how many copies there are of each block.

Or possibly:
select
 ct, count(*)
from
 (
 select file#, block#, count(*) ct
 from v$bh
 where objd = {data object id of table}
 group by
  file#, block#
 )
group by ct
/

To get a complete distribution pattern of how heavily cloned the blocks are.

The queries might be a little brutal on your buffer cache for several seconds, though.

Regards

Jonathan Lewis
http://jonathanlewis.wordpress.com

Author: Cost Based Oracle: Fundamentals
http://www.jlcomp.demon.co.uk/cbo_book/ind_book.html

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

• Original Message ----- -

> a. block cleanouts could cause some blocks to be updated (I think
> that would be a worst case of doubling.)

--
http://www.freelists.org/webpage/oracle-l