X-Sylpheed-Account-Id:1
S:andy.sharp@onstor.com
SCF:#mh/Mailbox/sent
X-Sylpheed-Sign:0
X-Sylpheed-Encrypt:0
X-Sylpheed-Privacy-System:
RMID:#imap/andys@onstor.net@onstor-exch02.onstor.net/INBOX	0	BB375AF679D4A34E9CA8DFA650E2B04E0AF1E040@onstor-exch02.onstor.net
X-Sylpheed-End-Special-Headers: 1
Date: Thu, 17 Jul 2008 13:32:37 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: "Rich LaReau" <rich.lareau@onstor.com>
Cc: "dl-cstech" <dl-cstech@onstor.com>
Subject: Re: Optimal IP configs
Message-ID: <20080717133237.115286f9@ripper.onstor.net>
References: <BB375AF679D4A34E9CA8DFA650E2B04E0AF1E040@onstor-exch02.onstor.net>
Organization: Onstor
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

On Thu, 17 Jul 2008 12:54:36 -0700 "Rich LaReau"
<rich.lareau@onstor.com> wrote:

> 
> We need to migrate about 6T of data from one cluster to another, and
> will be using client-based copies over NFS shares.  
> 
> My question, is there any advantage to setting up multiple virtual
> servers each with a unique IP address to virtually "split the load"?
> Or is having one vsvr with all the volumes configured with a single IP
> address on an aggregated port going to result in exactly the same
> throughput?

I would say that depends on the characteristics of the data.  Mostly a
large collection of smallish files, like, say, home directories, or a
collection of relatively large files like movies?

Most likely the storage will be the slowest part of the equation.  In
general, having multiple clients will increase the randomness of the
I/O from the array's viewpoint, only slowing things down further. Using
a single client, possibly running something like rsync (with the
--inplace option for a slight speed increase) or two tar processes, with
aggregated ports, would probably be the fastest.
