Date: Mon, 21 Feb 2000 18:55:15 -0600 From: "Jeffrey J. Mountin" <jeff-ml@mountin.net> To: Cy Schubert - ITSD Open Systems Group <Cy.Schubert@uumail.gov.bc.ca> Cc: stable@FreeBSD.ORG Subject: Re: Initial performance testing w/ postmark & softupdates... Message-ID: <3.0.3.32.20000221185515.00a2f880@207.227.119.2> In-Reply-To: <200002211524.HAA53892@cwsys.cwsent.com> References: <Your message of "Sat, 19 Feb 2000 03:53:42 CST." <3.0.3.32.20000219035342.009ce460@207.227.119.2>
next in thread | previous in thread | raw e-mail | index | archive | help
At 07:24 AM 2/21/00 -0800, Cy Schubert - ITSD Open Systems Group wrote: >A paper by Gregory R. Granger and Yale N. Patt of the Departement of >EECS, University of Michigan entitled Soft Update: A Solution to the >Metadata Update Problem in File Systems discusses softupdate >performance v.s. asynchronous write performance. The difference wasn't >much, somewhere around 2-3%. Only recently enabled softupdates. Used an asynch /usr/obj (striped vinum volume) for building and noatime on several mounts. Vividly recall the problems way back when with softupdates and never got around to testing it. It has been mature for a while, so the change was overdue. My /usr/obj are^H^H^Hwere a couple old 540 Quantum IDE drives. Very noisy when thrashed by buildworld/installworld. Had to retire one that was developing bad sectors a few weeks back. Now with the single running softupdates it is no longer thrashed quite as hard. Loss of the striped volume hasn't changed by going from noatime+asynch to noatime w/softupdates. Gained a couple minutes (few %) for buildworlds, but I'll find another pair of newer old drives and swap them in. Jeff Mountin - jeff@mountin.net Systems/Network Administrator FreeBSD - the power to serve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3.0.3.32.20000221185515.00a2f880>