CCL:Gaussian, Linda, and PC clusters

[eugene.leitl@lrz.uni-muenchen.de]

(((Please direct your comments/suggestions to the original poster below)))

From: dominic ditoro <dditoro@hydroqual.com>
To: "'chemistry@ccl.net'" <chemistry@ccl.net>
Subject: CCL:Gaussian, Linda, and PC clusters
Date: Fri, 11 Feb 2000 20:50:10 -0500

We have ~10k$ to spend in the near future at which point it evaporates... We
are considering purchasing a cluster of fast PC's, say 4 machines initially
(700 Mhz or so, 256Mb, 8 GB), to run Gaussian using linux and Linda. Has
anyone had any experience with this sort of configuration. We guess that it
will be roughly 5-7 times faster than our present single 400 Mhz machines.
We don't have the resources to endlessly baby sit the cluster. So the ease
of use is as important as performance. Our vision is that compiled code will
arrive from Gaussian, we install the appropriate linux os, load the code and
start computing. Is this living in dreamland?

Finally, any experience with running highlevel methods, e.g. G2, on a
cluster?

Either public or private responses are welcome. I will summerize.


Dominic M. Di Toro
Professor
Environmental Engineering
Manhattan College, Riverdale NY
dditoro@hydroqual.com

-= This is automatically added to each message by mailing script =-
CHEMISTRY@ccl.net -- To Everybody    |   CHEMISTRY-REQUEST@ccl.net -- To Admins
MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH
CHEMISTRY-SEARCH@ccl.net -- archive search    |    Gopher: gopher.ccl.net 70
Ftp: ftp.ccl.net  |  WWW: http://www.ccl.net/chemistry/   | Jan: jkl@osc.edu




-------------------------------------------------------------------
To unsubscribe send a message body containing "unsubscribe"
to beowulf-request@beowulf.org