From owner-soc-status@FreeBSD.ORG Mon Jul 6 14:56:33 2009 Return-Path: Delivered-To: soc-status@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 07AAF106564A for ; Mon, 6 Jul 2009 14:56:33 +0000 (UTC) (envelope-from fabio@freebsd.org) Received: from sssup.it (ms01.sssup.it [193.205.80.99]) by mx1.freebsd.org (Postfix) with ESMTP id 9300D8FC1D for ; Mon, 6 Jul 2009 14:56:31 +0000 (UTC) (envelope-from fabio@freebsd.org) Received: from [193.205.82.7] (HELO gandalf.sssup.it) by sssup.it (CommuniGate Pro SMTP 4.1.8) with ESMTP-TLS id 52075706 for soc-status@freebsd.org; Mon, 06 Jul 2009 16:44:31 +0200 Received: from smaug.retis (smaug.retis [10.30.3.72]) by gandalf.sssup.it (8.12.10/8.12.10) with ESMTP id n66EuUiT009927 for ; Mon, 6 Jul 2009 16:56:30 +0200 Received: by smaug.retis (Postfix, from userid 1000) id BA1F3538C3; Mon, 6 Jul 2009 16:57:39 +0200 (CEST) Date: Mon, 6 Jul 2009 16:57:39 +0200 From: Fabio Checconi To: soc-status@freebsd.org Message-ID: <20090706145739.GA14563@gandalf.sssup.it> References: <20090607105508.GS19347@gandalf.sssup.it> <20090622151559.GK28770@gandalf.sssup.it> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090622151559.GK28770@gandalf.sssup.it> User-Agent: Mutt/1.4.2.3i Subject: gsched - status report X-BeenThere: soc-status@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Summer of Code Status Reports and Discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 06 Jul 2009 14:56:33 -0000 Hi all, during the last two weeks I've updated the code to use the bio classification machinery commited by luigi@ in HEAD, and I've added some basic tracking of threads' behavior to drive the heuristics controlling some of the scheduling parameters. The first two milestones are almost complete, even if the proxy destruction code still needs more testing with ongoing I/O. About the third (and last), benchmarking, I've been playing around with microbenchmarks (using fio, http://git.kernel.dk/?p=fio.git;a=summary) and I've started looking at sysbench for its oltp implementation and at TPCC-UVa, an open source tpc-c implementation.