Thanks, Jared.

Yes, running select vs select+rollback in a loop of
1000 iterations I got similar results (average over 3
runs):

STAT...user rollbacks      0  1,000  1,000
LATCH...enqueues         570  1,574  1,004
LATCH...shared pool    7,434  9,063  1,629
STAT...recursive       7,754 10,264  2,510
LATCH...library cache 13,401 17,429  4,028

So, every select w/ a rollback is at least 1 enqueue
and 4 latches (1 shared pool + 3 library cache) more
expensive than select wo/ a rollback.

Since latches are scalability inhibitors - would it be
far away from the truth to say that an application
that uses OCI session pooling (as opposed to
connection pooling I suppose? - just reading chapter 9
of the OCI manual) doesn't scale well with respect to
the number of concurrent users?

Thanks,
Boris Dali.

 --- Jared Still <[EMAIL PROTECTED]> wrote: > Using a
slightly modified version of run_stats to
> return
> timings in 1/10000 of a second, the timing of 1000
> iterations
> of a loop executing noop vs. 1000 doing rollback:
> 
> .0005 secs
> .0354 secs
> 
> Here are the stats that were different between the
> two,
> kind of what you would expect:
> 
> LATCH.cache buffers chains                      113 
>       133        20
> STAT...redo size                              27160 
>     27648       488
> STAT...user rollbacks                             0 
>      1000      1000
> LATCH.enqueues                                    0 
>      1001      1001
> LATCH.shared pool                                 3 
>      1004      1001
> LATCH.session idle bit                            0 
>      1001      1001
> STAT...execute count                              3 
>      1004      1001
> LATCH.library cache pin                          13 
>      2014      2001
> LATCH.session allocation                          0 
>      2002      2002
> STAT...recursive calls                            4 
>      2008      2004
> LATCH.library cache                              14 
>      3018      3004

______________________________________________________________________ 
Post your free ad now! http://personals.yahoo.ca
-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.net
-- 
Author: Boris Dali
  INET: [EMAIL PROTECTED]

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).

Reply via email to