Date: Fri, 4 Dec 2009 15:22:39 -0500 From: Maxim Khitrov <mkhitrov@gmail.com> To: Peter Steele <psteele@maxiscale.com> Cc: "freebsd-questions@freebsd.org" <freebsd-questions@freebsd.org> Subject: Re: Is there the equivalent of a Windows "fast format" for UFS? Message-ID: <26ddd1750912041222v2e468e61i1800ef8d385606fd@mail.gmail.com> In-Reply-To: <7B9397B189EB6E46A5EE7B4C8A4BB7CB33C6C072@MBX03.exg5.exghost.com> References: <7B9397B189EB6E46A5EE7B4C8A4BB7CB33C6C072@MBX03.exg5.exghost.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Dec 4, 2009 at 2:23 PM, Peter Steele <psteele@maxiscale.com> wrote: > I suspect I know the answer to this question but I'll ask it anyway. We're dealing with some very large disks (11TB raid array) and a newfs operation takes a significant time. Is there any way to get a volume formatted faster than the typical newfs does? > It's actually the other way around - there is only "fast" format for UFS. As far as I know, full format for FAT or NTFS also checks for bad sectors on the drive, so that's why Windows has that distinction. What you need to do is properly configure block/frag sizes and inode density. Figure out the average file size that will be stored on the volume and divide the total volume size by that number. That is the minimum number of inodes you should create, since one inode is required per file. Of course, you should aim higher since you will not be able to add more inodes without reformatting the system. Increasing block and fragment sizes may also help, depending on what type of files you are storing. Read tuning(7) and then newfs(8) for options -b, -f, and -i. - Max
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?26ddd1750912041222v2e468e61i1800ef8d385606fd>