Subject: Re: Raidframe experiments and scsi woes
To: Michael VanLoon <mvanloon@MindBender.serv.net>
From: Thor Lancelot Simon <tls@panix.com>
List: current-users
Date: 11/26/1998 23:35:50
On Thu, Nov 26, 1998 at 01:34:27PM -0800, Michael VanLoon wrote:
> RAID-5 doesn't give you high performance.  I think to get high
> performance *and* high reliability, you need to forget about parity, and
> do normal RAID-0 striping on top of pairs of RAID-1 mirrored drives.
> This gives you full ccd-like striped write performance, and double read
> performance since it can interleave reads to between the mirrored
> drives.

FWIW, on SCSI<->SCSI RAID controllers with hardware parity support (e.g. 
the CMD 5500) RAID 5 is not significantly slower than RAID 4, and faster,
of course, for small writes.

RAID 4 is probably the best answer, *if* one has a filesystem that can
arrange to always, or almost always, write exactly the segment size.  LFS
could do this, if it worked...

Another answer is RAID 4 with a mirrored/striped parity disk.