Re: cache buffers chain latch contention

  • From: "Jonathan Lewis" <jonathan@xxxxxxxxxxxxxxxxxx>
  • To: <oracle-l@xxxxxxxxxxxxx>
  • Date: Thu, 25 Mar 2004 19:44:58 -0000

Notes in-line

Regards

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

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

April 2004 Iceland  http://www.index.is/oracleday.php 
June  2004      UK - Optimising Oracle Seminar


----- Original Message ----- 
: From: "Khedr, Waleed" <Waleed.Khedr@xxxxxxx>
: To: <oracle-l@xxxxxxxxxxxxx>
: Sent: Thursday, March 25, 2004 7:08 PM
: Subject: RE: cache buffers chain latch contention
:

 :Degree of 8 means 16 workers on 2 CPUs = means magnifying any small
: contention for CPU resources (that will never show up on another system
: if you have more CPU's).

    Not necessarily 16 slaves with degree 8.
    In fact, the path for this query shows no GROUP BY,
    ORDER BY, hash or merge joins, and particularly
    no Parallel to Parallel: so it will use only 8 slaves.

    Depending on the actual activity, 8 slaves is just
    about supportable on 2 CPUs, although that
    does depend on a reasonable split of disk and
    CPU activity.  In this case, degree 4 might be
    more sensible.

: Reduce the degree to 2 and see how it works for you.



----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to:  oracle-l-request@xxxxxxxxxxxxx
put 'unsubscribe' in the subject line.
--
Archives are at //www.freelists.org/archives/oracle-l/
FAQ is at //www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------

Other related posts: