Date: Tue, 3 Oct 1995 10:41:11 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: gibbs@freefall.FreeBSD.org (Justin T. Gibbs) Cc: terry@lambert.org, gibbs@freefall.FreeBSD.org, rsnow@txdirect.net, nc@ai.net, questions@freebsd.org Subject: Re: RAID solutions for FreeBSD Message-ID: <199510031741.KAA01408@phaeton.artisoft.com> In-Reply-To: <199510030332.UAA07368@aslan.cdrom.com> from "Justin T. Gibbs" at Oct 2, 95 08:32:55 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> >I *know* that 1-3 can be done without sequencer level coding. I suspect
> >that 4-5 can be too, albiet at a *very* high cost in processing time.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> >Check out the ZEBRA FS papers (RAID II).
>
> I read it last year. It would be a total waste to do the parity
^^^^^^
> calculations on the host processor when the card will do this for
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> you (assuming I can get docs).
I know, I know -- it's just that the idea of a software-only soloution
is very attractive. Putting the calculations in the sequencer code
may be more efficient, but it's wholly unportable.
Well, unless we write a high level sequencer language and "compile" it
to NCR + Adaptec, etc. I don't know how similar the sequencer instruction
set is; probably not very.
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199510031741.KAA01408>