3.543 supercomputing the humanities, cont. (64)

Willard McCarty (MCCARTY@vm.epas.utoronto.ca)
Thu, 5 Oct 89 18:43:37 EDT

Humanist Discussion Group, Vol. 3, No. 543. Thursday, 5 Oct 1989.


(1) Date: Thu, 5 Oct 89 02:18:46 EDT (23 lines)
From: "Patrick W. Conner" <U47C2@WVNVM>
Subject: 3.541 supercomputing the humanities? (46)

(2) Date: Thu, 05 Oct 89 09:53:57 PLT (20 lines)
From: "Guy L. Pace" <PACE@WSUVM1>
Subject: Supercomputing in the humanities

(1) --------------------------------------------------------------------
Date: Thu, 5 Oct 89 02:18:46 EDT
From: "Patrick W. Conner" <U47C2@WVNVM>
Subject: 3.541 supercomputing the humanities? (46)

I took a course in supercomputing such as David Stueler describes,
because I thought supercomputing might be the next frontier in humanistic
computing. What I discovered was that I didn't have the time or
expertise to prepare the CRAY to do what I'd like to have done. I think
that, for us, the promise of supercomputing lies in the tremendous
ammount of memory available and speed of access, but to make use of those,
we need a body of texts available to the machine unlike any yet
available. What we need is a supercomputer with access to all of the
texts to be named in the Rutger's Inventory project, wherever they
may be. Then, using the supercomputer, you define the texts which
interest you, and you ask the machine to generate the data from that set
of texts. After that, I suspect that you would download the generated
data to a local mainframe and continue your project in the manner
appropriate to the study. I may be all wet, but I think that those
of us who are not willing to become supercomputing engineers will have
to wait for the engineers to get around to setting things up for us,
just as we did (and are doing) with pre-super computers.
-
--Pat Conner
(2) --------------------------------------------------------------24----
Date: Thu, 05 Oct 89 09:53:57 PLT
From: "Guy L. Pace" <PACE@WSUVM1>
Subject: Supercomputing in the humanities

In response to David Stuehler's request for information on the John von
Neumann National Supercomputer Center, I have just a few comments and
recommendations. The center has four supercomputers and four VAX 8600's
(two on VMS and two on Ultrix). So, as far as computing power is concerned
there is plenty.
As a potential user of the center, your first question should be: What
applications are available on the supercomputers?
What applications are available should help you decide whether the center
can meet your computing needs.
JvNC does have user consultants, applications consultants and research
scientists on staff. If the applications support your "non-science"
programs, contact one of the applications consultants and discuss specific
tasks and research.
I suspect the "hard science" folks will get the most from using the
supercomputer center. Putting a text data base in a supercomputer seems
equivalent to using the M1 ABRAMS tank for swatting flies.