AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:<20081017153111.75212bd3@ripper.onstor.net>
CFG:
PT:0
S:andy.sharp@onstor.com
RQ:
SSV:onstor-exch02.onstor.net
NSV:
SSH:
R:<narayan.venkat@onstor.com>
MAID:1
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
X-Sylpheed-End-Special-Headers: 1
Date: Fri, 17 Oct 2008 15:31:36 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: Narayan Venkat <narayan.venkat@onstor.com>
Subject: project M notes
Message-ID: <20081017153136.4fc1e15d@ripper.onstor.net>
Organization: Onstor
X-Mailer: Sylpheed-Claws 2.6.0 (GTK+ 2.8.20; x86_64-pc-linux-gnu)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

Hi N,

I spoke with Pat and it's now clear to me that for the immediate
future, a 1U + JBOD is probably our best choice for hardware.  Our
other option is probably 3U with 14-16 drive bays, but I'm guessing
that will be more money than 1U head + 2U-12 JBOD, plus it will
probably make customers happier because they can buy just one JBOD and
two heads.  Perhaps in the future there will be more 2U-12 options that
satisfy our requirements.

If we get head and JBODs from different vendors, we will have to get
matching faceplate made for one of the boxes, I'm OK with that.  Don't
fear the couple extra $$.  Negus' group can handle the job of getting
that accomplished, and I'm sure the overall result will be still less
$$ than the 3U chassis with dual ported drives.  The right supplier
might throw it in for free if we get Bob to do the negotiating.  Or you
or whoever is good at that.

We need to do some testing on various hardware configurations to
determine what is required CPU and memory wise.  To do that we need
access to some equipment on the Cats side of the street, namely a low
end cougar or a bobcat with not-too-high end storage attached so that
we can compare.  Either that or we need to know how to generate the
benchmark results that you call out in the PRD so we can tell when
we're hitting the marks.  I found out that Pat has not done any testing
with clients, so we really have no useable numbers.  Perhaps I can
arrange some testing to be done up in Pleasanton next week with
clients.

Cheers,

a
