Cluster computing software survey

Robert G. Brown rgb at phy.duke.edu
Wed Oct 31 06:07:40 PST 2001


On Tue, 30 Oct 2001, Walter Bischof wrote:

> Greg,
> 
> We are a number of students, trying to see what is the Beowulf
> community's needs and wants.  Sure, we can make the survey results
> available later.

That's fine, and I just sent back a survey.  However, I noted that you
forgot to put AMD processors down at all on the form.  Oops.  AMD's
processors and motherboards have somewhat dominated the list discussion
for the last four or five months and we certainly have a slew of them
here already and are getting more.  You also forgot to include the GSL
(Gnu Scientific Library) on your list of libraries.  I've been using it
through the last few betas (from 0.8-something up) and it is getting to
be quite lovely and functional.

Although I'm sure that many folks on the list use non-open source
scientific libraries out of necessity, I'm equally sure that just about
all of us would PREFER open source libraries for the usual reasons --
the ability to fix things that don't work, the ability to alter a core
e.g. ODE solver to work on odd packed matrices of equations instead of
vectors, the ability to actually LOOK at things like how special
functions are being evaluated so you can know whether or not to trust an
e.g. incomplete gamma function evaluated in some specific regime.
Numerical libraries are almost always built on top of openly published
algorithms and I've always thought that it is especially silly for them
to be sold in closed source binary form.

Just a bit of editorial opinion there, sorry;-)

    rgb

> 
> 
> W
> 
> 
> On Tue, 30 Oct 2001, Greg Lindahl wrote:
> 
> > Date: Tue, 30 Oct 2001 22:28:35 -0500
> > From: Greg Lindahl <lindahl at conservativecomputer.com>
> > To: beowulf at beowulf.org
> > Subject: Re: Cluster computing software survey
> >
> > On Tue, Oct 30, 2001 at 06:29:49PM -0700, Walter Bischof wrote:
> >
> > > We are conducting a survey on software in the cluster computing
> > > environment.  We would really appreciate it if you could please
> > spend a
> > > few minutes to answer the following questions, and send your
> > answers
> > > back to bischof at tiscali.co.uk.
> >
> > Who are you affiliated with, and do you plan on publishing the
> > results
> > of the survey? It is traditional for Internet surveys to publish
> > their
> > results.
> >
> > greg
> >
> > _______________________________________________
> > Beowulf mailing list, Beowulf at beowulf.org
> > To change your subscription (digest mode or unsubscribe) visit
> > http://www.beo$
> >
> 
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
> 

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu






More information about the Beowulf mailing list