From owner-freebsd-hardware Sun Aug 8 4: 5:54 1999 Delivered-To: freebsd-hardware@freebsd.org Received: from zed.ludd.luth.se (zed.ludd.luth.se [130.240.16.33]) by hub.freebsd.org (Postfix) with ESMTP id AD28F14E50 for ; Sun, 8 Aug 1999 04:05:47 -0700 (PDT) (envelope-from pantzer@speedy.ludd.luth.se) Received: from speedy.ludd.luth.se (pantzer@speedy.ludd.luth.se [130.240.16.164]) by zed.ludd.luth.se (8.8.5/8.8.5) with ESMTP id NAA28183; Sun, 8 Aug 1999 13:04:10 +0200 Message-Id: <199908081104.NAA28183@zed.ludd.luth.se> X-Mailer: exmh version 2.0.1 12/23/97 To: brian@pobox.com Cc: freebsd-hardware@FreeBSD.ORG Subject: Re: The $500 Performance Question In-Reply-To: Message from Brian McGroarty of "Fri, 06 Aug 1999 22:15:05 PDT." <19990807051505.3696.rocketmail@web1003.mail.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Date: Sun, 08 Aug 1999 13:04:10 +0200 From: Mattias Pantzare Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I'm not sure I see where building objects to another drive would > help either with lazy writes/soft updates on since the entire > block of files to be written easily fits within RAM. They are not cached forever. Dirty blocks are written everey 30 seconds. = Why = not just try a diffrent disk for obj? > Interestingly, despite there only being a (roughly) 15% decrease > in time, CPU usage jumps tremendously. Most of the time I'm > averaging about 20% free on one CPU and as little as 5% on the > other. I'll be curious to look at the soft update code - I > wonder if that's coded as efficiently as it could be. You can probably blame the giant locks instead. The CPUs will wait for = eachother a lot. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message