[Beowulf] Re: Grants for Beowulf Clusters

[Bjorn Tore Sund]

On Mon, 14 Mar 2005 beowulf-request at beowulf.org wrote:

> Message: 4
> Date: Mon, 14 Mar 2005 07:04:53 -0800
> From: "Jim Lux" <james.p.lux at jpl.nasa.gov>
> Subject: Re: [Beowulf] Grants for Beowulf Clusters
> To: "Timo Mechler" <mechti01 at luther.edu>, <beowulf at beowulf.org>
> Message-ID: <003401c528a7$2db91180$30f49580 at LAPTOP152422>
> Content-Type: text/plain;	charset="iso-8859-1"
> 
> I can't claim that I am successful at getting grants for clusters,
> however...
> If you can make a good case that a cluster will make it possible to solve
> some other "important" problem, the odds go up greatly.  Think of a cluster
> as a tool, just like a microscope or an ultra centrifuge or a furnace.  How
> would you justify getting the budget for a big microscope (like a SEM)?
> 
> The key is to have a problem that everyone wants to attack, and the cluster
> being the way to attack it.  You said you've been doing proof of concept..
> Is that to prove that you can build a cluster, or that you've demonstrated
> some useful "work" with the cluster on a problem that someone is interested
> in (i.e. for which there is funding available).
> 
> Otherwise, you're a solution looking for a problem.

Moving further away from the original question, I'd like to expand on 
the above.  A Beowulf cluster can be used to solve problems.  You need
the problem.  You also need to make sure that any and all clusters you
get have the necessary usage volume to warrant the purchase.

Assume you've successfully argued/proved that the problem you're trying
to solve can be addressed by using a Beowulf cluster.  Do you have the
human resources to actually do so?  There's been several cases of people
getting clusters to address specific problems, only to discover they 
don't have time to learn how to code properly for them, and nor does
anyone else.

And if your funding source is then of a type that frowns on their money
being spent with nothing really happening as a result, finding funding
for more useful stuff later isn't going to be easy.

Bjørn
-- 
Bjørn Tore Sund           Phone:  (+47) 555-84894    Stupidity is like a
System administrator      Fax:    (+47) 555-89672    fractal; universal and
Math. Department          Mobile: (+47) 918 68075    infinitely repetitive.
University of Bergen      VIP:    81724
Support: system at mi.uib.no Contact: teknisk at mi.uib.no Direct: bjornts at mi.uib.no