From oracle-l-bounce@freelists.org Fri May 20 15:14:39 2005 Return-Path: Received: from air891.startdedicated.com (root@localhost) by orafaq.com (8.12.10/8.12.10) with ESMTP id j4KKEdfA029108 for ; Fri, 20 May 2005 15:14:39 -0500 X-ClientAddr: 206.53.239.180 Received: from turing.freelists.org (freelists-180.iquest.net [206.53.239.180]) by air891.startdedicated.com (8.12.10/8.12.10) with ESMTP id j4KKEd4Z029104 for ; Fri, 20 May 2005 15:14:39 -0500 Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id D5AD319574C; Fri, 20 May 2005 14:11:46 -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 27136-04; Fri, 20 May 2005 14:11:46 -0500 (EST) Received: from turing (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 5F0EB195623; Fri, 20 May 2005 14:11:46 -0500 (EST) Message-ID: <20050520190957.72665.qmail@web52601.mail.yahoo.com> Date: Fri, 20 May 2005 12:09:57 -0700 (PDT) From: Martic Zoran Subject: Re: 10g System statistics - single and multi To: kutrovsky.oracle@gmail.com, breitliw@centrexcc.com Cc: Oracle-L@freelists.org In-Reply-To: <52a152eb05052011252d525a2a@mail.gmail.com> MIME-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit X-archive-position: 20040 X-ecartis-version: Ecartis v1.0.0 Sender: oracle-l-bounce@freelists.org Errors-To: oracle-l-bounce@freelists.org X-original-sender: zoran_martic@yahoo.com Precedence: normal Reply-To: zoran_martic@yahoo.com X-list: oracle-l X-Virus-Scanned: by amavisd-new-20030616-p9 (Debian) at avenirtech.net X-Spam-Level: X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on air891.startdedicated.com X-Spam-Status: No, hits=0.5 required=5.0 tests=AWL,FORGED_YAHOO_RCVD autolearn=no version=2.63 Christo, You should separate things that are maybe happening on the disk level or disk cache level from the Oracle optimizer COST. Oracle does not recognize is the read random disk I/O read or eventually sequential single disk read. For Oracle all single block reads are measured equally if not proved differently. What is happening after that at execution time neither Oracle or OS or I/O subsystem will give you measures, prediction or cost in front at the time COST is calculated. Also, not sure how are you going to prove anything about the difference while ORacle does not know anything about that difference or you have statistics/events recognizing or counting the difference. At the end, even Oracle developers are not building the neural network optimizer that will give you unpredictable execution plans because at the end you need to balance the gain of having simple deterministic (eve based on 100 of input parameters) useful algorithm and too complex non deterministic beast. Regards, Zoran Martic __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- http://www.freelists.org/webpage/oracle-l