NA Digest Sunday, April 25, 1999 Volume 99 : Issue 17

Today's Editor:
Cleve Moler
The MathWorks, Inc.

Submissions for NA Digest:

Mail to

Information about NA-NET:

Mail to

URL for the World Wide Web:

From: Gene Golub <golub@sccm.Stanford.EDU>
Date: Sun, 18 Apr 99 9:11:39 PDT
Subject: Roger Hockney

I'm very sad to report the death of Roger Hockney on April 14. Roger
was one of the leading experts on the use of parallel computers for
solving large scale scientific problems. He was one of the original
contributors to the Fast Poisson technology. Roger was deeply
interested in many scientific problems. As a human, he was very kindly
and thoughtful of others; I'm happy to have been his friend.



From: Garry Tee <>
Date: Mon, 19 Apr 1999 10:47:14 +1200
Subject: Roger Hockney


I was saddened to read Gene Golub's announcement that Roger Hockney
has died. He & I were mathematicians together from 1958 to 1964 at English
Electric Co. at Whetstone. We had frequent discussions about our work,
especially on solution of Poisson problems.
Roger spent much time on modelling a thermonuclear plasma,
computing the plane trajectories of about 100 ions & electrons in a
magnetic field. In about 1962 he expessed to me his concern about how could
make he any use of the thick piles of numerical printout generated by his
program, & I suggested that he make an animated cartoon.
Our laboratory at Whetstone was then equipped with a cathode-ray
tube, with a camera for movie film. Roger modified his programs to produce
maps of the ions and electrons,and he produced a short film (40 seconds, I
think) which vividly portrayed the motions of the ions and electrons. Roger
screened that film at many research establishments (including Livermore
Laboratories), and that made many researchers realize the usefulness of
animated cartoons for displaying dynamic systems.

Garry Tee,
Department of Mathematics, University of Auckland, Auckland, New Zealand


From: Gil Strang <>
Date: Sat, 24 Apr 1999 10:53:56 -0400 (EDT)
Subject: Gian-Carlo Rota

I regret very much to write that Professor Gian-Carlo Rota
of MIT passed away at his home at the age of 66 from heart failure.
He was a wonderful person -- so civilized, friendly, and encouraging.
And he was an outstanding mathematician, who began as an analyst
and went on to transform the subject of combinatorics.

If you ever read his articles and book reviews, or heard him
lecture, you would never forget him.

Gil (Gilbert Strang)


From: Michele Benzi <>
Date: Thu, 22 Apr 1999 09:39:30 -0600 (MDT)
Subject: Gian-Carlo Rota

GIAN-CARLO ROTA (1932-1999)

Gian-Carlo Rota, one of the leading mathematicians of our time, died
unexpectedly on Sunday, April 18 in Cambridge, MA. He was born in
Vigevano, Italy, on April 27, 1932. He was supposed to deliver the
prestigious "Grosswald Lectures" at Temple University in Philadelphia
on April 19-21.

After leaving Italy with his family in 1945, Rota lived in Ecuador
where he attended the American School of Quito. He came to the US
in 1950 to study at Princeton University, where he received a
B.A. (summa cum laude) in 1953. He continued his studies at Yale
University, where he received a Ph.D. in 1956 under Jacob T. Schwartz.

After holding positions at the Courant Institute and at Harvard, in
1959 Rota joined the faculty at the Massachusetts Institute of Technology
where he remained for the rest of his life, except for a two-year stint
as a professor at Rockefeller University in 1965-1967. Since 1975
Rota was Professor of Applied Mathematics and Philosophy at MIT.
During his tenure at MIT he supervised nearly 50 doctoral students.

In addition, Gian-Carlo Rota held visiting positions at a number of
universities over the years and was a consultant for the Rand Corporation
(1966-1971), Brookhaven National Laboratory (1969-1973), and Los Alamos
National Laboratory (1966-1999).

A member of the US National Academy of Sciences since 1982 and of
numerous other academies worldwide, Gian-Carlo Rota was the
recipient of several honorary degrees (including one in Computer
Science from the University of Bologna in Italy) and of the 1988
Steele Prize of the American Mathematical Society. In 1992 the
National Security Agency awarded Rota the Medal for Distinguished

Immensely knowledgeable in several fields of pure and applied mathematics
and an expert in phenomenology and existential philosophy, Gian-Carlo Rota
was best known for his fundamental contributions to combinatorics, invariant
theory and probability. The readers of NA Digest may be interested to know
that Rota, through his connection with the Los Alamos laboratory, was also
keenly interested in numerical analysis and scientific computing, and
a strong advocate and supporter of research in these areas at the lab.
One fruit of this interest is the volume "A history of computing in the
twentieth century", that Rota edited (together with Nick Metropolis and Jack
Howlett) for Academic Press in 1980.

In addition to his technical contributions, Gian-Carlo Rota was a prolific
essayist on the cultural aspects of mathematics---historical, biographical,
anecdotal, etc.---and the author of numerous book reviews. From these highly
enjoyable pieces one can form an idea of the breadth and depth of his culture
and get a taste of his charm and brilliant wit. Rota was a man of high standard
and strong opinions, which occasionally led him to peaks of devastating
abrasiveness, as exemplified for instance in some of his aphorisms and book
reviews. In spite of this rather formidable reputation, he was, above all, a
true gentleman. He will be sorely missed.

For further information, see

Michele Benzi
Scientific Computing Group, CIC-19
Los Alamos National Laboratory
Los Alamos, New Mexico (USA)


From: Gene Golub <golub@sccm.Stanford.EDU>
Date: Thu, 22 Apr 99 8:22:11 PDT
Subject: David Young's Thesis

David Young's thesis was one of the monumental works of modern
numerical analysis. The thesis has been Latexed (with a few edits)
and is available on the WEB.

You can retrieve it as follows:

Thanks, David, it is a remarkable contribution and has had enormous
impact for the computational and engineering communities.

Gene Golub


From: Dirk Laurie <>
Date: Mon, 19 Apr 1999 08:55:00 +0200 (SAT)
Subject: Bibliography on Computing Gaussian Formulas

Dear N.A. Netters,

I am writing a survey paper on the computation of Gaussian and
related quadrature formulas, i.e. any one-dimensional quadrature
formula of higher degree than interpolatory. I was struck by how
extremely convenient it is to have available the exhaustive
bibliography in:

author = Gautschi,
title = "A Survey of {G}auss-{C}hristoffel Quadrature Formulae",
booktitle = "{E. B. Christoffel}: The Influence of His Work on
Mathematics and the Physical Sciences", publisher = Birkhauser,
year = "1981", pages = "72-147"}

In order that my survey paper might one day also be convenient
to someone else, I would like to update Gautschi's bibliography
to cover the last two decades. May I ask N.A. Digest readers
to send me bibliographic details of any papers on the topic,
published 1980 and later, that they are aware of?

The e-mail address to use is:

At the bottom of this message I also give snailmail and fax

I shall thank by name in the paper everyone that sends me
a reference that I include, unless the sender expressly asks me
not to do so.

The idiosyncratic way in which I select the topics actually treated
in the body of the paper will remain my own responsibility and will
in no way be blamed on N.A. Digest readers.

Dirk Laurie, School for Modelling Sciences
Potchefstroom University for Christian Higher Education
P.O. Box 1174, Vanderbijlpark 1900, South Africa Fax:(27)(16)910-3614


From: Marcin Paprzycki <>
Date: Tue, 20 Apr 1999 00:34:33 -0500 (CDT)
Subject: New Book, High Performance Algorithms for Structured Matrix Problems

Edited by: Peter Arbenz, Marcin Paprzycki, Ahmed H. Sameh, Vivek Sarin

Volume II in the Series:
Advances in the Theory of Computation and
Computational Mathematics
ISBN 1-56072-594-X

The main aim of this volume is to summarize the state of the art in the
area of high performance solutions of structured linear systems as well as
the area of structured eigenvalue and singular-value problems. The volume
highlights research directions perceived to be the most important for
computing the structured problems.

The topics covered range from parallel solvers for sparse or banded linear
systems to parallel computation of eigenvalues and singular values of
tridiagonal and bidiagonal matrices. In addition, the volume contains
articles on specialized solution techniques for dense Toeplitz and Hankel

The papers also discuss implementation issues on numerous parallel
architectures such as vector computers, shared and distributed memory
multiprocessors, and clusters of workstations.

Contents (abstracts will be posted soon on the ACTP WWW site):

P. Arbenz (ETH Zurich), M. Paprzycki (University of Southern Mississippi),
A. Sameh and V. Sarin (Purdue University)

Part I
Linear System Solvers

A Comparison of Frontal Software with other Harwell Subroutine Library
Sparse DIrect Solvers
I.F. Duff and J.A. Scott (Rutherford Appleton Laboratory)

Sparse Matrix Bandwidth Reduction: Algoithms, Applications and Real
Industrial Cases in Electromagnetics
A. Esposito, M.S.F. Catalano, F. Malucelli and L. Tarricone (University of

On the Stable Parallel Solution of General Narrow Banded Linear Systems
P. Arbenz (ETH Zurich) and M. Hegland (Australian National University)

Part II
Eigenvalue Problems

Efficient Algorithms for Reducing Banded Matrices to Bidiagonal and
Tridiagonal Form
B. Lang (Bergische University Wuppertal)

Parallel bisection algorithms for solving the symmetric tridiagonal
J. M. Bad=EDa (University Jaume I. Castell=F3n) and A. M. Vidal (Technical
University Valencia)

A Parallel OR Algorithm for the Symmetric Tridiagonal Eigenvlue Problem
I. Bar-On (Technion University)

Part III
Matrices with Special Structure

A Numerical Cmparison of Look-Ahead Levinson and Schur ALgorithms for
Non-Hermitian Toeplitz Systems
M. Hochbruck (University of Tubingen)

Superfast Solution of Linear Equations with Low Displacement Rank
T. Huckle (Technical University Munich)

Part IV
Parallel Computation

Load Balance in Parallel FACR
L.S. Johnsson and N.P. Pitsianis (University of Houston)

Parallel CG-Methods - Automaticallu Optimized for PC- and Workstation
J. Eisenbiegler, J. Gottlieb, W. Lowe, S. Schlaeger, M Thul and W.
Zimmermann (University of Karlsruhe)


From: Stephen Kirkup <>
Date: Wed, 21 Apr 1999 10:23:30 +0100
Subject: New Book, The Boundary Element Method in Acoustics

New book: "The Boundary Element Method in Acoustics"
and Fortran Codes
Author: Stephen Kirkup
ISBN: 0 9534031 0 6

The boundary element method (BEM) is a powerful tool in computational
acoustic analysis. The Boundary Element Method in Acoustics serves as
an introduction to the BEM and its application to acoustic problems
and goes on to complete the development of computational models.
Software inmplementing the methods is available.

Boundary element methods are developed for three important
classes of acoustic or Helmholtz problem: modelling the acoustic
field either interior or exterior to a closed surface or carrying out
acoustic modal analysis. Each class of problem is considered in
two-dimensional, three-dimensional and axisymmetric cases.
A very general development of the methods is followed so that
the widest range of engineering problems is covered.

Fortran 77 codes implementing the methods are available. Test
problems are used to demonstrate the methods. Examples
of realistic applications such as vehicle noise and loudspeaker
design analysis illustrate the potential of the BEM in acoustic

Chapter 1: Introduction
Chapter 2: Boundary Representation
Chapter 3: The Discrete Helmholtz Operators
Chapter 4: The Interior Acoustic Problem
Chapter 5: The Exterior Acoustic Problem
Chapter 6: Interior Modal Analysis

More information (including price/how to order) can be found on the
web page

A review appears in the current Applied Mechanics Reviews,
Review 3R8, vol 52 no 3, March 1999, ppB22-B23.


From: Alberto Valli <>
Date: Thu, 22 Apr 1999 08:10:54 +0100
Subject: New Book, Domain Decomposition Methods for PDEs

Dear Sirs,

I would like to announce the publication of the book:

Alfio Quarteroni, Alberto Valli
Domain Decomposition Methods for Partial Differential Equations
xvi+360 pp, 35 line figures
Oxford University Press
ISBN 0-19-850178-1, =A3 55.00


1. The Mathematical Foundation of Domain Decomposition Methods
2. Discretised Equations and Domain Decomposition Methods
3. Iterative Domain Decomposition Methods at the Discrete Level
4. Convergence Analysis for Iterative Domain Decomposition Algorithms
5. Other Boundary Value Problems
6. Advection--Diffusion Equations
7. Time-Dependent Problems
8. Heterogeneous Domain Decomposition Methods
9. Appendix


Domain decomposition methods are designed to allow the effective numerical
solution of partial differential equations on parallel computer
architectures. They comprise a relatively new field of study, but have
already found applications in many branches of physics and engineering. In
this book the authors illustrate the basic mathematical concepts behind
domain decomposition, looking at a large variety of boundary value

Contents include: symmetric elliptic equations; advection--diffusion
equations; the elasticity problem; the Stokes problem for incompressible
and compressible fluids; the time-harmonic Maxwell equations; parabolic and
hyperbolic equations; and suitable couplings of heterogeneous equations.

Thank you for your cooperation. Best regards

Alberto Valli


From: Fred Kus <fred@blas.cis.McMaster.CA>
Date: Thu, 22 Apr 1999 11:42:49 -0400
Subject: Documentation for Cray FFT Routines


I need to port a program which calls the Cray-based FFT routines
RFFTMLT and CFFTMLT to a Sparc. Does anyone have documentation,
e.g. man pages, for these routines ?

Thanks very much.


Fred W. Kus INTERNET: fred@McMaster.CA
Computing & Information PHONE: (905) 525-9140 ext. 24160
Services FAX (905) 528-3773
McMaster University URL
Hamilton, Canada L8S 4M1 A. N. Bourns Bldg. Rm 131C


From: Erling Andersen <>
Date: Sat, 24 Apr 1999 20:46:14 +0200
Subject: MOSEK Optimization Package and Toolbox


For free downloads and the on-line manual see:

MOSEK is a software package for solving small-scale and large-scale
linear and convex optimization problems. The packages includes:

- An interior-point based optimizer using a homogeneous and self-dual model.
- An implementation of the primal simplex algorithm.
- A basis identification procedure for linear programs.
- An extensive presolve procedure for linear and nonlinear problems.
- An Application Program Interface (callable library).

And it reads and writes MPS files.

The MOSEK BASE SYSTEM solves the problems:
- linear programs (basic solution is always available)
- convex quadratic programs
- convex quadratically constrained programs
- second order cone programs

- linear programs (basic solution always available)
- convex quadratic programs
- convex quadratically constrained programs
- second order cone programs
- entropy programs
- geometric programs (posynomial case)
- linear least squares

In addition the toolbox can be used to read and write MPS files quickly.

Questions can be directed to:

Enjoy, Erling D. Andersen


From: Tony Chan <>
Date: Wed, 21 Apr 1999 14:29:14 -0700 (PDT)
Subject: Math Awareness Month: Mathematics and Biology

Dear NA-Digest,
April is Mathematics Awareness Month and the theme this year is
Mathematics and Biology. The UCLA Math and BioMath Depts have organized
the following lectures around this theme:

20/4 "Gene Mapping", Prof. Ken Lange, UCLA Dept. of Biomath and Genetics.
29/4 "Positron Emission Tomography: How to use math & physics to image
the biology of human disease", Michael Phelps, UCLA's Dept of Molecular
and Medical Pharmacology.
3/5 "Transition from Physical to Life Sciences", Dr. John Quackenbush, TIGR.
"TBA", Dr. Polly Moore, Genentech.
6/5 "What is Proteomics?", Prof. David Eisenberg, UCLA Dept. of Chemistry.
10/5 "The Age of a Disease Gene", Prof. Simon Tavare, USC Dept. of Math.

For more details, see

Tony Chan
Chair, UCLA Math Dept


From: Constructive Approximation <>
Date: Sun, 18 Apr 1999 16:43:23 -0400 (EDT)
Subject: Workshop on Minimum Energy Problems

A workshop on Minimum Energy Problems will be held Nov. 8-12,
1999 at the City University of Hong Kong as part of a series of
FoCM (Foundations of Computational Mathematics) workshops to be
held this fall.
The meeting will emphasize research work related to distributing
points on a sphere and on general Riemann surfaces, discrepancy
results (e.g. for Fekete points), potential theoretic tools and
applications to orthogonal polynomials, random matrices, integrable
systems, etc.
For further information, contact Ed Saff ( or
Arno Kuijlaars (


From: Gloream <>
Date: Tue, 20 Apr 1999 18:34:44 +0100 (GDT)
Subject: Global and Regional Atmospheric Modelling

Third GLOREAM Workshop
September 22-24, 1999
Ischia (Naples) - ITALY

Organized by: Center for Research on Parallel Computing
and Supercomputers (CPS-CNR), Naples, Italy

Co-sponsors: Naval University of Naples
University of Naples "Federico II"

The 1999 annual meeting of the GLOREAM (GLObal and REgional Atmospheric
Modelling) project will be held in Ischia (Italy) on September 22-24, 1999.
The meeting follows the first two GLOREAM workshops that were held in
Aachen (Germany) in September 1997 and in Madrid (Spain) in September 1998.

GLOREAM is a subproject of EUROTRAC-2, the second phase of the EUREKA
project on the transport and chemical transformation of environmentally
relevant trace constituents in the troposphere over Europe.
All topics related to long-range transport, global and regional modelling,
computational aspects, model validation and model application and
assessment studies are relevant for this workshop, including related
studies into the impact of chemical mechanisms, clouds and heterogeneous
processes, emission modelling, etc..

ALL RESEARCH GROUPS that have a project within GLOREAM are expected
to participate in this workshop and to give a presentation on their
work. However, also other researchers are invited to present results of their
work, if related to the workshop topics. The meeting will include also a
poster session.

Regina Isabella Hotel, island of Ischia, off the coast of Naples, Italy.

- Deadline for submission of abstracts: June 15, 1999
- Notification of acceptance: July 1, 1999
- Deadline for registration: July 15, 1999

210 Euro. It includes coffee breaks, lunches, a gala dinner, a booklet of
abstracts and a volume of proceedings.

Peter Builtjes (TNO, Apeldoorn, The Netherlands, co-ordinator of the project)
Michael Memmesheimer (University of Cologne, Cologne, Germany)
Hans Feichter (MPI, Hamburg, Germany)
Guido Barone (CPS-CNR & University of Naples "Federico II", Naples, Italy)
Giulio Giunta (CPS-CNR & Naval University of Naples, Naples, Italy)

Pasqua D'Ambra (CPS-CNR, Naples, Italy)
Angelo Ricco (CPS-CNR & University of Naples "Federico II", Naples, Italy)

at the following address:

Center for Research on Parallel Computing and Supercomputers
Via Cintia, Monte S. Angelo, 80126 Naples - Italy
FAX: +39-81-675636

To submit a contribution send a one-page abstract, in Word or LaTeX format,
to the e-mail address:

Further information are available at the following WEB SITE:


From: E. F. Toro <>
Date: Wed, 21 Apr 1999 09:06:19 +0100
Subject: Godunov Conference, Oxford UK


October 18th to 22nd, 1999

An international conference to review four decades of
research on Godunov and related methods, and to honour
Professor S. K. Godunov, the year of his 70th birthday


Professor S. K. Godunov Professor R. Abgrall
Dr H. Aiso Professor P. Colella
Dr. D. Drikakis Professor P. Garcia-Navarro
Professor B. Gustafsson Dr B. Koren
Professor D. Kroener Professor A. Marquina
Professor S. Osher Professor P. A. Raviart
Professor P. L. Roe Professor E. Romensky
Professor T. Saito Professor R. Saurel
Dr P. K. Sweby Professor E. Tadmor
Professor E. F. Toro Dr. I. Toumi
Professor L. N. Trefethen Professor B. Wendroff



Prospective contributors are invited to submit an abstract of no more than
one page in Latex (plus .ps files if any) no later than 30th June 1999 to: or post hard copy to:
Professor E. F. Toro
Department of Computing and Mathematics
Manchester Metropolitan University
Chester Street
Manchester, United Kingdom

Papers accepted for presentation will be considered for inclusion in the
book of proceedings, to be published in two volumes.


Details on the registration procedure for both the Conference and the
Short Course preceding the conference are given on the websites: and
or can be obtained by contacting: (Conference Organiser)


DERA Fort Halstead, UK NUMERITEK Limited, UK
Springer-Verlag, Germany John Wiley and Sons, UK
Manchester Metropolitan University, UK

Professor Dr E. F. Toro
Computational and Applied Mathematics Group
Department of Computing and Mathematics
Manchester Metropolitan University
Chester Street
Manchester M1 5GD

Telephone: (+44)161 247 3593; Fax: (+44)161 247 1483
E-Mail: (work); (home)


From: Raimondas Ciegis <>
Date: Wed, 21 Apr 1999 11:55:09 +0200 (METDST)
Subject: Conference in Lithuania on Mathematical Modelling and Analysis

Fourth International Conference
June 3-4, 1999, Vilnius, Lithuania

Sponsored by Lithuanian State Science and Studies Foundation


Conference organizers
- Institute of Mathematics and Informatics, Vilnius
- Vilnius Gediminas Technical University
- Vilnius University

The MMA conferences were established in 1996 to provide forum for
the discussion of various aspects of mathematical modelling
and usage of difference methods for numerical solution of modern
problems of science and engineering. Among the aims of the Conference
is the stimulation of the cooperation among practitioners and theoreticians
in this field.

Working language of the Conference is English.

The main topics of the Conference include:
- Analysis of numerical methods for solving problems of mathematical physics
- Parallel algorithms and parallel computing
- Application of difference methods to engineering problems
- Analysis of ODE and PDE problems and applications
- Statistical modelling
- Mathematical modelling for financial analysis
- Conceptual modelling

Program and Organizing Committee
R.Ciegis (Lithuania - General Chairman),
H.Neunzert(Germany - General Vice - chairman),
M.Sapagovas (Lithuania - the Chairman of the Program Committee),
A.Buikis (Latvia Vice-chairman), V.Abrashin (Belarus), S.Dey (USA),
F.Ivanauskas (Lithuania), H.Kalis (Latvia), A.Kjellman(Sweden),
P.Matus(Belarus), A.Reinfelds (Latvia), A.Pedas (Estonia), L.Saulis
(Lithuania), V.Skakauskas (Lithuania), A.Zemitis (Germany),
J.Wasniewski (Denmark).

The scientific Program includes invited plenary talks (40 min) and
contributed talks (20 min). Special sessions will be held on
selected topics of great current interest.

Correspondence address

Mathematical Modelling and Analysis - 99
Institute of Mathematics and Informatics,
Akademijos 4, LT-2600,
Vilnius, Lithuania

phone: (+370) 2 72 94 19, 2 72 94 17
fax: (+370) 2 72 92 09

Questions regarding MMA99 should be addressed to

Looking forward to meet you in Vilnius !

Prof. Raimondas Ciegis


From: George Miel <>
Date: Thu, 22 Apr 1999 23:50:16 -0700 (PDT)
Subject: Visiting Faculty Position at University of Nevada, Las Vegas

Applications are invited for a Visiting Assistant Professor for one
academic year beginning Fall 1999 in the

Department of Mathematical Sciences
University of Nevada, Las Vegas
Box 454020
Las Vegas, NV 89154-4020
(702) 895-3567

Minimum requirements include a Ph.D. in numerical analysis or a closely
related field with experience in scientific computing. Please send a
letter of application, vita, and at least three letters of recommendation
to one of the individuals listed below. E-mail submissions are encouraged.

Prof. Rohan Dalpatadu or Prof. George Miel

Review of applications will begin immediately. The position will remain
open until filled and is contingent upon funding. Salary will be commensurate
with qualifications and experience. For more information, see the UNLV World
Wide Web site at: AA/EOE. Minorities, Women, Veterans
and the Disabled are encouraged to apply.


From: Jim Blue <>
Date: Fri, 23 Apr 1999 15:31:50 -0400
Subject: Research Positions at NIST, Boulder

The Information Technology Laboratory of the National Institute of Standards
and Technology expects to have one or more vacancies later this year in the
area of

Mathematical Modeling and Computational Science

at its Boulder, Colorado campus. This position will be in the Mathematical
Modeling Group of the Mathematical and Computational Sciences Division.
Current research in this group includes modeling in the areas of
optoelectronics, electromagnetics (including materials characterization and
antenna metrology), cryocooler design, and wave propagation in elastic media.
Much of the group's work involves large scale scientific computation,
requiring the solution of partial differential equations and integral
equations. Expertise in analysis-based fast algorithms, such as the fast
multipole method and related methods, and high-order convergent
discretizations, for both differential and integral equations, is desirable
and would provide opportunity for collaboration within the group.

The candidate selected will initiate research collaborations with physicists,
chemical engineers, electrical engineers, or materials scientists in other
laboratories at NIST in Boulder (areas include modeling properties of
composites, material defects, chemical kinetics, fluid mechanics,
electromagnetic and acoustic wave propagation). The candidate will also
develop new numerical algorithms as required for these problems.

Minimum requirements are a Ph.D. (or equivalent experience) in applied
mathematics or a related field plus substantial knowledge of computational
physics, chemistry, or engineering; alternatively, a Ph.D. (or equivalent
experience) in a physical science with substantial knowledge of applied
mathematics and numerical methods. U.S. CITIZENSHIP IS REQUIRED.

Selective factors in this appointment will include:

1) Demonstrated experience in large-scale computational solution of problems
in physics, chemistry, or engineering.

2) Broad knowledge of applied mathematics, numerical algorithms, scientific
computing, and several areas of physics, chemistry, or engineering.

3) Broad research interests and demonstrated ability to collaborate with
scientists in several areas of physics, chemistry, or engineering.

4) Demonstrated ability to produce high quality mathematical software for
broad distribution.

Potential applicants are invited to send a resume, the names of three
references, and a brief description of research interests, addressing the
selective factors described above. Electronic submission is preferred, in
Ascii, Latex, TeX, or PostScript format, to Regular mail
submissions should be sent to
Dr. James L. Blue
100 Bureau Drive, Stop 8910
National Institute of Standards and Technology
Gaithersburg, MD 20899-8910

Expressions of interest must be received by May 31, 1999.

The National Institute of Standards and Technology is a committed Equal
Opportunity/Affirmative Action Employer, and strives to build a diverse
workforce by soliciting applications from women and members of minority


From: Wei Cai <>
Date: Sat, 24 Apr 1999 10:41:29 -0400 (EDT)
Subject: Visiting Faculty Position at UNC Charlotte

University of North Carolina at Charlotte
Department of Mathematics

Application is sought for a visiting assistant professor
position. Candidates are expected to have a Ph.D. in numerical
analysis and extensive experience in scientific computations.
Please e-mail application to


End of NA Digest