From owner-freebsd-hackers@FreeBSD.ORG Wed Sep 20 08:45:39 2006 Return-Path: X-Original-To: freebsd-hackers@freebsd.org Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 448DD16A407 for ; Wed, 20 Sep 2006 08:45:39 +0000 (UTC) (envelope-from marck@rinet.ru) Received: from woozle.rinet.ru (woozle.rinet.ru [195.54.192.68]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9003D43D53 for ; Wed, 20 Sep 2006 08:45:38 +0000 (GMT) (envelope-from marck@rinet.ru) Received: from localhost (localhost [127.0.0.1]) by woozle.rinet.ru (8.13.8/8.13.6) with ESMTP id k8K8jbHw066525; Wed, 20 Sep 2006 12:45:37 +0400 (MSD) (envelope-from marck@rinet.ru) Date: Wed, 20 Sep 2006 12:45:37 +0400 (MSD) From: Dmitry Morozovsky To: Kris Kennaway In-Reply-To: <20060919173421.GA45928@xor.obsecurity.org> Message-ID: <20060920123940.W63482@woozle.rinet.ru> References: <200609141232.k8ECWTXj045191@lurza.secnetix.de> <20060919160511.T33371@woozle.rinet.ru> <20060919173421.GA45928@xor.obsecurity.org> X-NCC-RegID: ru.rinet X-OpenPGP-Key-ID: 6B691B03 MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-2.0.2 (woozle.rinet.ru [0.0.0.0]); Wed, 20 Sep 2006 12:45:37 +0400 (MSD) Cc: freebsd-hackers@freebsd.org Subject: Re: numbers don't lie ... X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Sep 2006 08:45:39 -0000 On Tue, 19 Sep 2006, Kris Kennaway wrote: KK> > OF> Because buildworld is I/O-bound on systems with sufficiently KK> > OF> fast processors. KK> > OF> KK> > OF> Try putting the contents of /usr/src into a RAM disk and KK> > OF> repeat the benchmark. The numbers might look a little KK> > OF> different then. Of course, you should have sufficient RAM KK> > OF> in the machines -- If they're going to swap to the disks, KK> > OF> your benchmark won't be happy. KK> > OF> KK> > OF> I think putting /usr/obj onto a RAM disk is _not_ necessary KK> > OF> because of soft-updates, so the processes shouldn't block KK> > OF> on writes. KK> > KK> > My experiments show that if you have enough memory to host radmdrive for KK> > /usr/src you'd better leave it for caching - there were no statistically KK> > meaningful performance difference, at least on machines with 1G+ RAM. KK> KK> Really? My measurements show the opposite (on a system with 16GB of KK> RAM). My last test on amd64/dualcore with 4G of RAM and -j4 shows (buildworld+buildkernel): ==> /tmp/buildlog <== 1996.45 real 3032.94 user 624.83 sys Script done on Tue Sep 19 14:44:54 2006 ==> /tmp/buildlog.md <== 1957.45 real 3033.93 user 585.78 sys Script done on Tue Sep 19 15:20:42 2006 Second one was with 512M/4k/512 swap-backed md, the former with /usr/src on the gmirror'ed pair of SATAs. Sincerely, D.Marck [DM5020, MCK-RIPE, DM3-RIPN] ------------------------------------------------------------------------ *** Dmitry Morozovsky --- D.Marck --- Wild Woozle --- marck@rinet.ru *** ------------------------------------------------------------------------