[Beowulf] [jak@uiuc.edu: [Xgrid] Re: megaFlopsper Dollar? real world requirements]

[Ben Mayer]

Jim Gray talked about this. I can't find the paper atm, but did find
an interview with ACM Queue.

http://www.acmqueue.org/modules.php?name=Content&pa=printer_friendly&pid=43&page=1

>From what I remember of the paper he was saying that buying a whole
machine, packing it with disks and then shipping it can in the long
run be much cheaper then buying large internet connections. The
numbers have changed since he wrote the paper but given the major
trends of how storage/bandwidth have been changing I think his points
are more valid now.

One has to remember that one persons application needs are different
from someone else's. This may work well for you, it may not, but in
the end it is an interesting approach.

Anyways check out the above article for more information strait from
the masters mouth.

Ben


On 5/17/05, Eugen Leitl <eugen at leitl.org> wrote:
> ----- Forwarded message from "Jay A. Kreibich" <jak at uiuc.edu> -----
> 
> From: "Jay A. Kreibich" <jak at uiuc.edu>
> Date: Sun, 15 May 2005 16:25:38 -0500
> To: Eugen Leitl <eugen at leitl.org>
> Cc: xgrid-users at lists.apple.com
> Subject: [Xgrid] Re: megaFlopsper Dollar? real world requirements
> Reply-To: jak at uiuc.edu
> User-Agent: Mutt/1.4.2.1i
> 
> On Sun, May 15, 2005 at 06:48:20PM +0200, Eugen Leitl scratched on the wall:
> 
> > Don't forget that the highest data rate available is FedEx'ing a box of
> > diskdrives.
> 
>  Although there is a grand history of comments like this, (I believe
>  the original quote was "Never underestimate the bandwidth of a
>  station wagon full of tapes") it often isn't actually true.  Sure, it
>  wins "door-to-door", but having a box of hard drives next to my
>  computer does not make the data usable.  For a valid look you need to
>  look at memory-to-memory transfer.  In other words, include the time
>  write the hard drives, disconnect them, mail them, reconnect them,
>  and read all the data.  This always kills most tape arguments, and
>  has slowly been eroding the hard drive argument.