NA Digest Sunday, January 31, 1993 Volume 93 : Issue 5

Today's Editor:

Cleve Moler
The MathWorks, Inc.

Submissions for NA Digest:

Mail to

Information about NA-NET:

Mail to


From: Ronald N. Bracewell <>
Date: Mon, 25 Jan 1993 11:54:00 -0800 (PST)
Subject: Oscar Buneman

Oscar Buneman, Professor Emeritus of Electrical Engineering, passed
away peacefully in his sleep at his home on January 24, 1993. In recent years
he worked regularly in his office, arriving from Los Altos Hills on his
bicycle. Following a rigorous early education in the classical gymnasium in
Hamburg and two years of university, he moved to Manchester where he
completed a further five years of studies in mathematics. During World War II
he worked on the magnetron with the British Admiralty, on the calutron in
Berkeley, and on reactor research in Canada. For ten years he was a lecturer
in mathematics at Cambridge University and a fellow of Peterhouse college,
working in close contact with D.R. Hartree and P.A.M. Dirac who were principal
influences in his development. As a visiting scientist at Stanford in 1957 he
found the climate congenial and from 1960 has been a member of the Electrical
Engineering department. As an emeritus, his door was always open and he was
most generous with help to colleagues and students. Professor Buneman was a
great outdoorsman and, with his wife Ruth, spent as much time in the mountains
and sun as possible. Until Christmas he was a familiar sight in very brief
shorts, a very deep tan and an ancient crash helmet, cruising across campus on
his racing bike. He will be much missed by his colleagues.


From: Stephane Gerbi <>
Date: Mon, 25 Jan 93 11:09:17 MET
Subject: Nonliear Parabolic PDE

Dear Na-netters
I would like to know if it exists an efficient program
(in Fortran if possible) or a numerical method easy to programm
to solve a 1D fully nonlinear parabolic equation of the form:

Ut = f(U, Ux, Uxx) in (O,L)

+ Neumann Boundary Conditions

The function f is very smooth but the equation is degenerate i.e:
it exists a time T0, and a point X0 where U tends to zero and Ut tends
to infinity: the solution quenchs.
I was thinking of finite differences + Newton method.
I will appreciate any help

Stephane GERBI
Ecole Normale Superieure de Lyon
Unite de Mathematiques Pures et Appliquees
46, Allee d'Italie
69364 Lyon Cedex 07

phone: +33
fax: +33


From: Bob Fennell <>
Date: 26 Jan 1993 21:04:42 -1812
Subject: Graduate Programs in Applied Mathematical Sciences

Conference on Graduate Programs in the Applied
Mathematical Sciences II
April 16-18 , 1993
Clemson University
Clemson, SC

Current challenges for graduate education in classical analysis,
computational and discrete mathematics, operations research and
statistics will be discussed at this conference hosted by the
Department of Mathematical Sciences at Clemson University.

Specific conference objectives are to review the current status of
graduate education in the applied mathematical sciences; to review
the role of recent developments in the mathematical sciences in
the solution of modern technological problems and the challenges
and opportunities these developments present to graduate
education; to obtain perspectives on current industrial/government
employment opportunities and placement; and to publicize
innovative graduate programs to the professional community.

Participants will be selected to provide broad representation from
all areas of the mathematical sciences, including individuals from
academic, governmental and industrial institutions. The program
will consist of a number of panel, audience, and group
discussions. This format necessitates a limited number of
participants. A conference proceedings will be published.

Preliminary Program
Friday April 16
I. Graduate Mathematical Sciences Education: An Historical Perspective.
II. Current Status of Applied Mathematical Sciences Graduate Education
in the U.S.
III. Challenges for Graduate Education in the Mathematical Sciences.

IV. Curricula for the 90's and Beyond.

V. Curriculum Innovations.

Saturday April 17

VI. Computation in the Curriculum.
VII. Employment Opportunities for the 90's.

VIII. Non-Academic Careers for MS and Ph.D. Students.
XI. Group Discussions.
1. Recruiting and Advising Graduate Students.
2. Department Administration, University Support, Rewards.
3. Undergraduate Preparation.

Sunday April 18
X. Summary and Closing Remarks.

Invited participants include: John Burns, Don Bushaw, Ron Douglas, James
Dyer, Bob Easterling, Avner Friedman, Carl Harris, Robert Hogg, Bill Lucas,
Robert Lundegard, Tom Magnanti, Joyce McLaughlin, Anna Nagurney,
Fred Roberts, Mary Wheeler, Doug Zahn.

This conference is supported by a grant from the National Security
Agency and additional funding from NSF is expected. For further
information contact, R. E. Fennell, Department of Mathematical
Sciences, Clemson University, Clemson SC 29634-1907, (803) 656-
3257, email


From: Donna DiLisi <>
Date: Wed, 27 Jan 93 09:22:19 EST
Subject: SIAM Student Travel Awards for 1993


During 1993, SIAM will make several awards of $250 to support student
travel to the following SIAM conferences:

Sixth Conference on Parallel Processing for Scientific Computing;
March 22-24 in Norfolk, VA

Conference on Mathematical and Computational Issues in
Geosciences; April 19-21 in Houston, TX

Conference on Mathematical and Numerical Aspects of Wave
Propagation; June 7-10 in Newark, DE

Annual Meeting; July 12-16 in Philadelphia, PA

Conference on Simulation and Monte Carlo Methods; August 4-6
in San Francisco, CA

Third Conference on Linear Algebra in Signals, Systems & Control;
August 16-19 in Seattle, WA

Third Conference on Geometric Design; November 1-5 in
Tempe, AZ

The awards are to be made from the SIAM Student Travel Fund, created in
1991 and maintained through book royalties donated by generous SIAM

Any full-time student in good standing is eligible to receive an
award. Top priority will be given to students presenting papers at
the meetings, with second priority to students who are co-authors of
papers to be presented at the meetings. Only students traveling more
than 100 miles to the meetings are eligible for the awards.

An application for atravel award should include: (1) a letter from the
student stating the meeting for which support is being requested; (2) a
letter from the student's adviser or department chair stating that the
applicant is a full-time student in good standing; (3) if applicable,
the titles of the papers to be presented (or co-authored) by the
student at the meetings.

Applications should be sent to the SIAM office (Attn: SIAM Student
Travel Award), 3600 University City Science Center, Philadelphia, PA
19104-2688. Students may also apply by e-mail to or by
fax at 215-386-7999, but the letter from an adviser or department
chair must be an original, sent by postal mail.

Applications (including supporting letters) must be received at the
SIAM office by no later than one month before the first day of the
meeting for which support is requested.

Winners will be notified by no later than two weeks before the first
day of the meeting. Checks for the awards will be given to winning
students when they register at the given mee ting.

For further information about these awards, please contact Allison
Bogardo or Donna DiLisi at the SIAM office.

SIAM, 3600 University City Science Center, Philadelphia, PA
19104-2688;; (215) 382-9800 (phone); (215)
386-7999 (fax)


From: Iain Duff <>
Date: Fri, 29 Jan 93 18:51:40 GMT
Subject: SPARse KERnel Project

SPARKER project

The SPARse KERnel project is currently primarily concerned with the
development of kernels that assist in the development and porting of
software for the iterative solution of sparse equations. A major aim
of this project is to establish standards to enable efficient, and
portable, implementations of iterative algorithms on high-performance
computers. We envisage that a major benefit of this project will
be a much quicker implementation of new algorithmic ideas into complex
application packages.

The first two reports in the SPARKER series are now available on
anonymous ftp.

These are

SPARKER Working note 1
A proposal for user level sparse BLAS.
I Duff, M Marrone, and G Radicati


SPARKER Working note 2
A proposal for a sparse BLAS toolkit.
M Heroux

We would very much welcome comments on these proposals from any readers of
the digest. Information on how to obtain the papers is given below.

Iain Duff (
Mike Heroux (
Michele Marrone (
Giuseppe Radicati (

To get SPARKER reports


When prompted for a userid, reply with


and give your email address as a password.

Then cd to pub/open

Set mode to binary

Report 1 is in
Report 2 is in


End of NA Digest