From owner-freebsd-arch Sun Feb 9 17:12:11 2003 Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BCF0137B401 for ; Sun, 9 Feb 2003 17:12:09 -0800 (PST) Received: from flood.ping.uio.no (flood.ping.uio.no [129.240.78.31]) by mx1.FreeBSD.org (Postfix) with ESMTP id D2C1A43F93 for ; Sun, 9 Feb 2003 17:12:06 -0800 (PST) (envelope-from des@ofug.org) Received: by flood.ping.uio.no (Postfix, from userid 2602) id 922E1536E; Mon, 10 Feb 2003 02:12:04 +0100 (CET) X-URL: http://www.ofug.org/~des/ X-Disclaimer: The views expressed in this message do not necessarily coincide with those of any organisation or company with which I am or have been affiliated. To: Poul-Henning Kamp Cc: arch@freebsd.org Subject: Re: Our lemming-syncer caught in the act. From: Dag-Erling Smorgrav Date: Mon, 10 Feb 2003 02:12:03 +0100 In-Reply-To: <31118.1044817404@critter.freebsd.dk> (Poul-Henning Kamp's message of "Sun, 09 Feb 2003 20:03:24 +0100") Message-ID: User-Agent: Gnus/5.090014 (Oort Gnus v0.14) Emacs/21.2 (i386--freebsd) References: <31118.1044817404@critter.freebsd.dk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Poul-Henning Kamp writes: > An image is worth a thousand words, but in this case it only > says three words: "Man, that sucks!". Yes, it's really amazing that you draw conclusions about the quality of our syncer (and expect us to instantly agree with them) based on such a shitty graph. I see no information at all about how it was created, what the units are for the axes (and the green curve is most likely in a different unit than the red bars), your bars are averages which means there's no information about how many requests were satisfied at each point in time, you don't say anything about what you call a request and how you determine when it has been satisfied, etc., etc. For all I know the green curve could show requests at the UFS level and the red bars could show the time it takes before they hit the disk, in which case the graph simply shows that softupdates is enabled on the partition you did the testing on. If you're trying to say what I think you're trying to say - i.e. "some requests take a long time to be satisfied" - you should plot a histogram where each bin corresponds to a time range and the height of each bar corresponds to the number of requests whose age was in that range when they were satisfied. And you should make very very sure that your definitions of "initiated" and "satisfied" were consistent (i.e. at the exact same level in the I/O stack). DES -- Dag-Erling Smorgrav - des@ofug.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message