From oracle-l-bounce@freelists.org Fri Jun 18 03:06:24 2004 Return-Path: Received: from air189.startdedicated.com (root@localhost) by orafaq.com (8.11.6/8.11.6) with ESMTP id i5I869O29124 for ; Fri, 18 Jun 2004 03:06:19 -0500 X-ClientAddr: 206.53.239.180 Received: from turing.freelists.org (freelists-180.iquest.net [206.53.239.180]) by air189.startdedicated.com (8.11.6/8.11.6) with ESMTP id i5I85x629092 for ; Fri, 18 Jun 2004 03:06:09 -0500 Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 648E372C0B2; Fri, 18 Jun 2004 02:50:18 -0500 (EST) Received: from turing.freelists.org ([127.0.0.1]) by localhost (turing [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 15310-58; Fri, 18 Jun 2004 02:50:18 -0500 (EST) Received: from turing (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id A6C4972C0C7; Fri, 18 Jun 2004 02:50:17 -0500 (EST) Received: with ECARTIS (v1.0.0; list oracle-l); Fri, 18 Jun 2004 02:48:55 -0500 (EST) X-Original-To: oracle-l@freelists.org Delivered-To: oracle-l@freelists.org Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id D0E1A72C08A for ; Fri, 18 Jun 2004 02:48:54 -0500 (EST) Received: from turing.freelists.org ([127.0.0.1]) by localhost (turing [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 15846-07 for ; Fri, 18 Jun 2004 02:48:54 -0500 (EST) Received: from amethyst.servlets.net (amethyst.servlets.net [209.162.192.13]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 28E2572C013 for ; Fri, 18 Jun 2004 02:48:54 -0500 (EST) Received: from danismentgu (212-133-142-2.sbs.net.tr [212.133.142.2]) by amethyst.servlets.net (8.12.8/8.12.8) with SMTP id i5I89S8H026307 for ; Fri, 18 Jun 2004 01:09:30 -0700 Message-ID: <04fa01c4550b$b2cff1b0$fe19100a@danismentgu> From: "Danisment Gazi Unal (ubTools)" To: References: Subject: Re: The best CPU usage measurement in Oracle: BUFFER_GETS or CPU_TIME? Date: Fri, 18 Jun 2004 11:10:08 +0300 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_04F3_01C45524.D0EA34D0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4927.1200 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4939.300 X-Virus-Scanned: by amavisd-new at freelists.org X-archive-position: 2986 X-ecartis-version: Ecartis v1.0.0 Sender: oracle-l-bounce@freelists.org Errors-To: oracle-l-bounce@freelists.org X-original-sender: dunal@ubTools.com Precedence: normal Reply-To: oracle-l@freelists.org X-list: oracle-l X-Virus-Scanned: by amavisd-new at freelists.org ------=_NextPart_000_04F3_01C45524.D0EA34D0 Content-Type: text/plain; charset="windows-1257" Content-Transfer-Encoding: quoted-printable Jurijs, There are still many traps in Oracle response time measurements. You can = see them at = http://www.ubtools.com/cgi-bin/ib/ikonboard.cgi?act=3DPages;page=3Dmrpp As a short glance, "don't ignore logial IO, but starts with CPU time". I'm developing some new formulas for response time analysis. Some of = them is related to measure impact of CPU power and logical IO on CPU = time. I guess they will be available soon in itrprof. best regards... http://www.ubTools.com Web Based Oracle Products and Services ----- Original Message -----=20 From: J.Velikanovs@alise.lv=20 To: oracle-l@freelists.org=20 Sent: Friday, June 18, 2004 10:47 AM Subject: The best CPU usage measurement in Oracle: BUFFER_GETS or = CPU_TIME? My regards to all members,=20 Just would like to know your opinion.=20 Since 9i (suppose 9.2) we have CPU_TIME column in the V$SQL view.=20 I wonder which figure is the best measurement of CPU usage BUFFER_GETS = or CPU_TIME?=20 = =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=20 Lets imagine I have system with CPU bottleneck (can see height =93load = average=94 from OS) for a 1-3 ours. No particular long sessions have = been executed. It is seams mainly OLTP system. Parse CPU usage not the = issue.=20 I would like to identify TOP CPU consumers. As we all know I the V$SQL = is the best information source in this case.=20 = =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=20 I wonder which figure is the best measurement of CPU usage: = BUFFER_GETS or CPU_TIME? Statspack report, as well as Anjo Kolk = www.oraperf.com recommending to look on BUFFER_GETS not to CPU_TIME. Is = it just tradition or there are some arguments not to look on CPU_TIME as = main CPU usage indicator.=20 One reason I can imagine why CPU_TIME better indicator then = BUFFER_GETS is sorting. I can imagine that BUFFER_GETS not taking in = account CPU spent to sorting staff. Then from CPU usage perspective = better indicator is CPU_TIME.=20 Please correct me if I am wrong.=20 Thanks in advance,=20 Jurijs ------=_NextPart_000_04F3_01C45524.D0EA34D0 Content-Type: text/html; charset="windows-1257" Content-Transfer-Encoding: quoted-printable
Jurijs,
 
There are still many = traps in=20 Oracle response time measurements. You can see them at http://www.ubtools.com/cgi-bin/ib/ikonboard.cgi?act=3DPages;page=3D= mrpp
 
As a short glance, "don't = ignore logial=20 IO, but starts with CPU time".
 
I'm developing some new = formulas for=20 response time analysis. Some of them is related to measure impact of CPU = power=20 and logical IO on CPU time. I guess they will be available soon in=20 itrprof.

best regards...
 
http://www.ubTools.com
Web Based = Oracle=20 Products and Services
----- Original Message -----
From:=20 J.Velikanovs@alise.lv
To: oracle-l@freelists.org
Sent: Friday, June 18, 2004 = 10:47=20 AM
Subject: The best CPU usage = measurement=20 in Oracle: BUFFER_GETS or CPU_TIME?


My regards to all=20 members,
Just would like = to know your=20 opinion.
Since 9i (suppose = 9.2) we=20 have CPU_TIME column in the V$SQL view.

I wonder which figure is the best measurement of CPU usage = BUFFER_GETS=20 or CPU_TIME?


=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=20
Lets imagine I have system with = CPU=20 bottleneck (can see height =93load average=94 from OS) for a 1-3 ours. = No=20 particular long sessions have been executed. It is seams mainly OLTP = system.=20 Parse CPU usage not the issue.

I=20 would like to identify TOP CPU consumers. As we all know I the V$SQL = is the=20 best information source in this case.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=20

I wonder which figure is the = best=20 measurement of CPU usage: BUFFER_GETS or CPU_TIME? Statspack report, = as well=20 as Anjo Kolk www.oraperf.com recommending to look on BUFFER_GETS not = to=20 CPU_TIME. Is it just tradition or there are some arguments not to look = on=20 CPU_TIME as main CPU usage indicator.

One reason I can imagine why CPU_TIME better indicator then = BUFFER_GETS=20 is sorting. I can imagine that BUFFER_GETS not taking in account CPU = spent to=20 sorting staff. Then from CPU usage perspective better indicator is=20 CPU_TIME.
Please correct = me if I am=20 wrong.

Thanks in = advance,=20
Jurijs ------=_NextPart_000_04F3_01C45524.D0EA34D0-- ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@freelists.org put 'unsubscribe' in the subject line. -- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------