Hi, On Thu, Apr 03, 2003 at 09:28:36AM +0200, Emile van Bergen wrote: > Partitioning your data across disks is important, but IMHO partitioning > a single disk is useless. Enforcing quota by splitting a disk in two, > with all the seek time it wastes, is an unreasonably expensive way to do > it. > > If your server has only a single disk, I say don't bother with > partitions other than swap. Having different filesystems with different > properties can be immensely useful, but then you should use multiple > disks. Something just occurred to me. A lot of systems will have one (logical) disk, either physical or as a RAID-5 or RAID-1 set. Wouldn't it be nice if you could interleave multiple filesystems on the same block device? I.e. instead of giving one filesystem blocks 0 through x and another filesystem x+1 through y, you'd decide how many fs'es you want interleaved, and then give one fs blocks 0 + xn, another 1 + xn and a third 2 + xn, up to n-1 + xn, where n is the number of filesystems and x the number of blocks per filesystem? That way, you can have multiple filestystems that are immune to each others corruptions, each with their own type or other properties, that can share a single logical disk, without the seek penalty. Also, each fs can benefit equally from the higher troughput at the beginning of an oversized disk. It's like RAID-0, but slightly different in that instead of one filesystem being striped across disks, you stripe multiple filesystems on the same logical volume. Does anybody know if this has been proposed before? It shouldn't be too hard to achieve on the md layer; instead of allowing just one md per group of disks, allow multiple striped md's per group of disks. Or am I being stupid, either because you don't gain anything or because it's already implemented (I don't use software RAID here)? Inquiring minds would like to know. Cheers, Emile. -- E-Advies - Emile van Bergen emile@e-advies.nl tel. +31 (0)70 3906153 http://www.e-advies.nl
Attachment:
pgppwlGnfqGVN.pgp
Description: PGP signature