Subject: Re: Raidframe experiments and scsi woes
To: Michael VanLoon <mvanloon@MindBender.serv.net>
From: Dave Sainty <dave@dtsp.co.nz>
List: current-users
Date: 11/28/1998 01:16:20
"Michael VanLoon" writes:

> 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.

Surely RAID-5 will give you similar performance to ccd on reads.  Only
write performance will suck :)

Your arrangement will give very high write performance, as a cost
tradeoff...

Whilst I'm here...  In NZ, SCSI drives are incredibly expensive
compared to IDE, and IDE performance seems to be really pretty good
now, especially with Manuel's excellent and much appreciated work on
NetBSD's IDE support (Thanks Manuel!).

Ignoring the annoying fact that PCI IDE cards seem awfully hard to
come by (it's always on the motherboard now), is seems like I can get
4 cheap IDE drives into my machine on the existing motherboard
support, RAID-5 them, and get a pretty good result for around half the
price of a similar SCSI system.  Considering this is for a machine
that rarely has anyone but me working on it, this seems like a
sensible way to go.

Anyone got good or bad comments about using IDE drives for medium
sized disk arrays?

Cheers,

Dave