Date: Tue, 14 Aug 2001 07:18:27 -0500 From: Mike Meyer <mwm@mired.org> To: Andrew J Caines <A.J.Caines@halplant.com> Cc: "Robert J. Adams" <radams@siscom.net>, freebsd-hackers@freebsd.org Subject: Re: Large /dev/md0 ramdisk Message-ID: <15225.5907.82100.155925@guru.mired.org> In-Reply-To: <20010814011705.D67874@hal9000.servehttp.com> References: <078a01c12466$345dc830$7215fbd1@jason> <20010814011705.D67874@hal9000.servehttp.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Andrew J Caines <A.J.Caines@halplant.com> types: > > From what I've read, md is a better way of going than mfs. > I have seen several references to md being preferred to mfs but have been > unable to find any detailed comparison or evaluation. To what reference > were you referring? md and mfs have two different behaviors. md uses real memory. mfs uses buffer space, which is backed by disk. Which is better depends on what you need it for. > I've also seen it suggested that mfs is to go away, leaving only md. For > happy mfs users like myself who enjoy the convenience of a simple flexible > fstab entry sufficing for memory backed non-persistent filesystems, that > is of some concern. md has been replaced/upgraded in -current, and now includes the functionality of both mfs and vn. If I understand things correctly, the new md implementation of mfs will perform better than mfs. There has been talk of a "make and mount" command for md that would work like mount_md, but it doesn't appear to have materialized yet. There are also patches floating around for /etc/rc that let you set a couple of things in /etc/rc.conf to get your md file system set up at boot. <mike -- Mike Meyer <mwm@mired.org> http://www.mired.org/home/mwm/ Independent WWW/Perforce/FreeBSD/Unix consultant, email for more information. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?15225.5907.82100.155925>