From owner-freebsd-performance@FreeBSD.ORG Sat Jan 26 20:00:32 2008 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6BBAF16A47E for ; Sat, 26 Jan 2008 20:00:32 +0000 (UTC) (envelope-from erik@cederstrand.dk) Received: from mx2.itu.dk (unknown [130.226.142.29]) by mx1.freebsd.org (Postfix) with ESMTP id 37D0E13C47E for ; Sat, 26 Jan 2008 20:00:32 +0000 (UTC) (envelope-from erik@cederstrand.dk) Received: from wimac.littlebit.dk (unknown [85.233.238.191]) by mx2.itu.dk (Postfix) with ESMTP id 0337AF48072 for ; Sat, 26 Jan 2008 21:00:30 +0100 (CET) Message-ID: <479B9147.1080801@cederstrand.dk> Date: Sat, 26 Jan 2008 21:00:07 +0100 From: Erik Cederstrand User-Agent: Thunderbird 2.0.0.9 (Macintosh/20071031) MIME-Version: 1.0 To: "freebsd-performance@freebsd.org" References: <4796C717.9000507@cederstrand.dk> In-Reply-To: <4796C717.9000507@cederstrand.dk> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: Performance Tracker project update X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 26 Jan 2008 20:00:32 -0000 Erik Cederstrand wrote: > > I'd like to send a small update on my progress on the Performance > Tracker project. > > I now have a small setup of a server and a slave chugging along, > currently collecting data. I'm following CURRENT and collecting results > from super-smack and unixbench. > > The project still needs some work, but there's a temporary web interface > to the data here: http://littlebit.dk:5000/plot/. Apart from the > plotting it's possible to compare two dates and see the files that have > changed. Error bars are 3*standard deviation, for the points with > multiple measurements. Per the suggestions in this thread, the web interface has now grown an overview page and an alerts page. The alerts are calculated as anything outside mean +/- 2*standard deviation in a sliding window of 10 dates. This was found by trial-and-error to catch most significant changes without picking up too much noise. Anything below a 1% change is discarded. Erik