Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: _serial_direct_read effect

Re: _serial_direct_read effect

From: Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk>
Date: Mon, 7 Feb 2005 07:51:13 +0000 (UTC)
Message-ID: <cu76lh$mnu$1@titan.btinternet.com>

One of the oddities about serial_direct_read is that it seems to become an attribute of the cursor. This means that you won't see a difference in your test because the second run of the SQL uses the cursor opened in the first part of the test, and therefore does not do direct reads. You need to have two slightly different statements if you want to see the effects in the stats and latches.

Your theory about less CPU because of the reduction of buffer and latch activity is correct - but the effect may not be significant (The same is true for a change in block size).

It is not 'good' to use the parameter, because it is a hidden parameter - and these, more or less by definition, are not good parameters to use on production systems without the direct approval of Oracle support.

Regards

Jonathan Lewis

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

http://www.jlcomp.demon.co.uk/seminar.html Public Appearances - schedule updated Jan 21st 2005

<agonenil_at_gmail.com> wrote in message
news:1107729611.453578.112930_at_c13g2000cwb.googlegroups.com...
> Hi ,
> I've a situation that I need to read a big table very fast.
> I'm using FTS, but I don't want to use parallel slave (not to over load
> the system). I need to read this data only once
>
> I've tried to play with _serial_direct_read but I've found any
> differnt.
> Alought I think it should reduce cpu usage and latches (not going on
> cache).
>
Received on Mon Feb 07 2005 - 01:51:13 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US