[Beowulf] Grants for Beowulf Clusters

[Andrew Leahy]

We had some success at getting funding from the NSF/CCLI program for a 
college-level beowulf cluster:

https://www.fastlane.nsf.gov/servlet/showaward?award=0089045

It might be a good idea to search through NSF Fastlane to see what else 
they've funded through CCLI.

I personally don't believe that a dedicated beowulf cluster is 
necessarily a good investment for a strictly undergraduate institution. 
  We may be outliers among liberal arts colleges, but many of our 
scientists aren't all that computationally-minded and so far our cluster 
hasn't seen any use outside of the mathematics and computer science 
departments.  However, our computer scientists were looking for a linux 
lab (on our exclusively Windows campus) for use in instruction.  So we 
argued that we could build a dual use linux lab/beowulf cluster that we 
could hold classes in and use as a cluster.

Technically, this isn't a beowulf cluster. However, we gave each system 
dual processors (reasoning that a student simply typing away on a 
document wouldn't really notice if one of the processors was working on 
a computationally intensive program at the same time) and we equipped 
each system with two network cards, one of which was hooked up to our 
own dedicated "beowulf" network.  So you might say it was "beowulf-like".

I can't say that our grant was a tremendous success, but that's largely 
for external reasons. We built it at the height of the ".com" era, when 
there were gobs of computer science students.  As it was originally 
envisioned, the course we designed would be a numerical analysis course 
with an emphasis on parallel algorithms (a sexy subject for CS 
types--there isn't anything else about distributed memory programming in 
the curriculum) primarily for solving large systems of linear equations. 
  However, when ".com" went bust CS enrollments dried up and we haven't 
really had a big audience.  Right now, I'm retooling the course to focus 
on applied partial differential equations, again with an emphasis on 
solving large systems of equations with parallel algorithms.  We'll see 
if it can pick up a broader audience among science students in general.

If anybody else has ideas for distributed computing topics that would go 
well in an undergraduate numerical analysis course, please let me know.

Andrew Leahy
Knox College

Timo Mechler wrote:
> Hi all,
> 
> I'm wondering what kind of success rate people are having with obtaining
> grants for Beowulf type Linux Clusters (for example, from the National
> Science Foundation).  Let me give you a little bit more info as to why I'm
> asking this:  I'm a junior undergraduate at a small liberal arts college
> in Iowa (~2600 students), and have solely been pursuing Beowulf clusters
> for well over a year now.  I believe strongly that even though that my
> school is small, several departments on campus could benefit from the use
> of a beowulf cluster in the research that does go on.  I've been using
> older, slower machines as a proof of concept for now.  Ideally, we would
> want a faster beowulf system eventually that offers significant
> improvements over anything desktop pc's have to offer nowadays.  Being
> that money is an issue at smaller schools, is there any I could obtain a
> grant for a beowulf cluster?  If so, besides the NSF, what would be some
> other sources to apply to?  Since some of you guys on this come from big
> companies or Univesities, I would appreciate any insight and suggestions
> you can give me.  All input is appreciated.  Thanks in advance!
> 
> Best Regards,
> 
> -Timo Mechler
> 

---
[This E-mail scanned for viruses by Declude Virus]