RE: Buffer use under Oracle
- From: mail@xxxxxxxxxxxxxxxxxxx
- To: oracle-l@xxxxxxxxxxxxx
- Date: Wed, 12 May 2004 20:05:27 +0100 (BST)
Is it in an oversized keep pool? And so the unnecessary
cloned blocks are not being aged out?
-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx]On Behalf Of Cary
Millsap
Sent: 12 May 2004 19:39
To: oracle-l@xxxxxxxxxxxxx
Subject: RE: Buffer use under Oracle
I didn't get the sense from the original note that
there was a 'free buffer
waits' problem, so I'm assuming that there's not really
demand for new
buffers. I thought the original post was an expression
of curiosity about
how there could be so many index blocks in the buffer
cache at the same
time. (Perhaps I didn't read carefully enough, and it's
of course difficult
to tell at this point what the original post said
because the original post
isn't present in this thread anymore.)
I'm curious though, if a cloned buffer is accessed
frequently by running
queries, are you saying that the probability is high
that it will get aged
out prematurely, and therefore force re-reconstruction
the next time that
particular incarnation of the buffer is required?
Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com
* Nullius in verba *
Upcoming events:
- Performance Diagnosis 101: 5/18 Edison NJ, 6/22
Pittsburgh, 7/20 Boston
- SQL Optimization 101: 5/24 San Diego, 6/14 Chicago,
6/28 Denver
- Hotsos Symposium 2005: March 6-10 Dallas
- Visit www.hotsos.com for schedule details...
-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx]
On Behalf Of K Gopalakrishnan
Sent: Wednesday, May 12, 2004 10:47 AM
To: oracle-l@xxxxxxxxxxxxx
Subject: Re: Buffer use under Oracle
Cary:
I don't think this is right unless I am reading between
lines ;).
Cloned buffers are always kept in the cold end in the
new alogorithm
and they are ready to be flushed when ever there is a
demand for
new buffers. The CR buffers are kept in the FROZEN end
(note the
word freeze in the parameter, it is not cold. Freeze)
Of course the behavior can be controlled by
_db_aging_freeze_cr
parameter.
KG
> >Why aren't the old ones freed when a new one is
created?
>
> ...because of Oracle's LRU buffer cache management
algorithm.
> Currently-executing queries may "like" having a
slightly older CR copy =
> of a
> newly modified block in the buffer cache. Such a
block will tend to =
> remain
> cached as long as some session keeps using it. Those
blocks are in your
> buffer cache because your application needs them.
>
----------------------------------------------------------------
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
-----------------------------------------------------------------
----------------------------------------------------------------
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
-----------------------------------------------------------------
----------------------------------------------------------------
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:
