From owner-freebsd-fs@FreeBSD.ORG Sun Jan 9 02:22:06 2011 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2EEC6106566C for ; Sun, 9 Jan 2011 02:22:06 +0000 (UTC) (envelope-from bfriesen@simple.dallas.tx.us) Received: from blade.simplesystems.org (blade.simplesystems.org [65.66.246.74]) by mx1.freebsd.org (Postfix) with ESMTP id E8B978FC0A for ; Sun, 9 Jan 2011 02:22:05 +0000 (UTC) Received: from freddy.simplesystems.org (freddy.simplesystems.org [65.66.246.65]) by blade.simplesystems.org (8.13.8+Sun/8.13.8) with ESMTP id p092M4mW007455; Sat, 8 Jan 2011 20:22:04 -0600 (CST) Date: Sat, 8 Jan 2011 20:22:04 -0600 (CST) From: Bob Friesenhahn X-X-Sender: bfriesen@freddy.simplesystems.org To: Ivan Voras In-Reply-To: Message-ID: References: User-Agent: Alpine 2.01 (GSO 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.2 (blade.simplesystems.org [65.66.246.90]); Sat, 08 Jan 2011 20:22:05 -0600 (CST) Cc: freebsd-fs@freebsd.org Subject: Re: Another Phoronix article X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 09 Jan 2011 02:22:06 -0000 On Sun, 9 Jan 2011, Ivan Voras wrote: > Phoronix is not exactly the epitome of correct benchmarking but it appears > that they've at least recently found out about the existence of error bars. Yes, but you go read the site anyway. > In summary: very varied results, some are expected (low parallel write small > IOPS for FreeBSD), some are not (apparently the BSDs have a monopoly on > high-performance gzip :) ). But overall, pretty good relative results for > FreeBSD, better than earlier. Most of the results don't seem very believable to me. The filesystems behave quite differently in terms of the filesystem coherency they assure. Filesystem caching can also be quite different. There is no telling what these various benchmarks are actually measuring. The only slightly trustable benchmark I saw was the time to extract the Linux kernel, and that has to be taken with a grain of salt since it is not clear what would be on disk if one was to pull the power plug immediately when the extraction was claimed to be done. In some cases, possibly almost nothing. Bob -- Bob Friesenhahn bfriesen@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/