From owner-freebsd-chat@FreeBSD.ORG Wed Sep 30 09:11:16 2009 Return-Path: Delivered-To: freebsd-chat@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id DAF2F106568B for ; Wed, 30 Sep 2009 09:11:16 +0000 (UTC) (envelope-from soralx@cydem.org) Received: from idcmail-mo2no.shaw.ca (idcmail-mo2no.shaw.ca [64.59.134.9]) by mx1.freebsd.org (Postfix) with ESMTP id B249C8FC25 for ; Wed, 30 Sep 2009 09:11:16 +0000 (UTC) Received: from pd5ml1no-ssvc.prod.shaw.ca ([10.0.153.166]) by pd6mo1no-svcs.prod.shaw.ca with ESMTP; 30 Sep 2009 02:42:55 -0600 X-Cloudmark-SP-Filtered: true X-Cloudmark-SP-Result: v=1.0 c=1 a=YK3RP7YmNK0A:10 a=0S1XfAVaquWMAfU7ho/htg==:17 a=-rGBIGOjGWobw4lHJkkA:9 a=18UMTjiLqSbhskqbFiMA:7 a=6ATd-nRqczyj6sAc9BIz0OQuTREA:4 Received: from unknown (HELO cydem.org) ([24.87.3.133]) by pd5ml1no-dmz.prod.shaw.ca with ESMTP; 30 Sep 2009 02:42:54 -0600 Received: from soralx (soralx [192.168.0.240]) by cydem.org (Postfix/FreeBSD) with ESMTP id ED7B97E804; Wed, 30 Sep 2009 01:42:56 -0700 (PDT) Date: Wed, 30 Sep 2009 01:42:51 -0700 From: To: Message-ID: <20090930014251.4f827302@soralx> In-Reply-To: References: <200909290226.CAA28246@sopwith.solgatos.com> X-Mailer: Claws Mail 3.7.2 (GTK+ 2.16.6; i386-portbld-freebsd7.2) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: freebsd@sopwith.solgatos.com, lists@stringsutils.com Subject: Re: FreeBSD vs Ubuntu - Discuss... X-BeenThere: freebsd-chat@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Non technical items related to the community List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 30 Sep 2009 09:11:16 -0000 > Moving to chat instead of performance. > > >> This was discussed in detail in slashdot.. starting with the fact that > >> most likely debug switches were not turned off for FreeBSD. > > > > "All of the FreeBSD and Ubuntu options were left at their defaults." > > > > My question is why is FreeBSD's disk i/o performance so bad? > > As I mentioned... this was discussed actively in slashdot. You will find > there many good comments on this. Debug switches? Irrelevant, as 7.2 performed just as poorly (if not worse) in the threaded random writes test. One would think that the unrealistically poor [disk?] I/O performance bench data in FreeBSD was just a glitch, but using the OS everyday as a workstation, I actually notice that there could be some truth in those numbers. At least for ATA, when there's some disk I/O going on, file write operations that normally take milliseconds, may take tens of seconds or a minute! For example, loading the root disk with some serious concurrent I/O (portupgrade, find, tar xz, etc) makes opera unusable: the web browser normally saves "sessions" file everytime there's a change (e.g., a tab closed, or a page scrolled), and usually the write operation is unnoticeable, but with heavy disk I/O, one could wait for tens of seconds before, say, a page gets scrolled following keyboard input. I thinks that stream [memory benchmark] may also be demonstrating a weakness in FreeBSD, though I have doubts on this one. --- [SorAlx] ridin' VN2000 Classic LT