Received: from localhost (maia-3.hub.org [200.46.204.184]) by postgresql.org (Postfix) with ESMTP id A1EEF9FB446 for ; Fri, 16 Feb 2007 21:44:39 -0400 (AST) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.184]) (amavisd-new, port 10024) with ESMTP id 88764-01 for ; Fri, 16 Feb 2007 21:44:34 -0400 (AST) X-Greylist: from auto-whitelisted by SQLgrey-1.7.4 Received: from floppy.pyrenet.fr (floppy.pyrenet.fr [194.116.145.2]) by postgresql.org (Postfix) with ESMTP id BD32C9FB2CB for ; Fri, 16 Feb 2007 21:44:38 -0400 (AST) Received: by floppy.pyrenet.fr (Postfix, from userid 106) id 4DDC830E2B; Sat, 17 Feb 2007 02:44:37 +0100 (MET) From: Ron Mayer X-Newsgroups: pgsql.general Subject: Re: Priorities for users or queries? Date: Fri, 16 Feb 2007 17:44:40 -0800 Organization: Hub.Org Networking Services Lines: 28 Message-ID: <45D65E08.7000306@cheapcomplexdevices.com> References: <000001c74da2$a9a4a990$6400a8c0@dualcore> <45CEE33B.8020303@hagander.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@news.hub.org Cc: Jan Wieck , Ron Mayer To: Magnus Hagander User-Agent: Icedove 1.5.0.9 (X11/20061220) In-Reply-To: <45CEE33B.8020303@hagander.net> To: pgsql-general@postgresql.org X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200702/920 X-Sequence-Number: 109410 Magnus Hagander wrote: > Most likely, you do not want to do this. You *can* do it, but you are > quite likely to suffer from priority inversion Papers I've read suggest that the benefits of priorities vastly outweigh the penalties of priority inversion for virtually all workloads on most all RDBMs's including PostgreSQL. This CMU paper in particular tested PostgreSQL (and DB2) on TPC-C and TPC-W workloads and found that indirectly influencing I/O scheduling through CPU priorities is a big win for postgresql. http://www.cs.cmu.edu/~bianca/icde04.pdf "For TPC-C running on PostgreSQL, the simplest CPU scheduling policy (CPU-Prio) provides a factor of 2 improvement for high-priority transactions, while adding priority inheritance (CPU-Prio-Inherit) provides a factor of 6 improvement while hardly penalizing low-priority transactions." Have you heard of any workload on any RDBMS where priority inversion causes more harm than benefit? Ron Mayer