## NA Digest Sunday, April 16, 2000 Volume 00 : Issue 16

Today's Editor:
Cleve Moler
The MathWorks, Inc.
moler@mathworks.com

### Submissions for NA Digest:

Mail to na.digest@na-net.ornl.gov.

### Information about NA-NET:

Mail to na.help@na-net.ornl.gov.

URL for the World Wide Web: http://www.netlib.org/na-net/na_home.html
-------------------------------------------------------

From: Vincent Picaud <picaud@ige.univ-fcomte.fr>
Date: Wed, 12 Apr 2000 15:37:17 +0200
Subject: Hamilton-Jacobi Equation

I'm PhD student using the level-set method. But I encountered the following
problem (presented here in 2D) which I think can interest people working
on numerical methods for the Hamilton-Jacobi equation.

Statement of the problem :

-> From the level-set method :

D[Phi,t] + F.Norm[Grad[Phi]]=0

I introduce the following speed F=tn.M.n where n is the
normal vector (n=Grad[Phi]/Norm[Grad[phi]]), tn
the transposition of this vector and M the 2x2 matrix below:

| vx 0 |
M= | 0 vy |

So, the level-set equation gives the following Hamilton-Jacobi
equation :

D[Phi,t] + H[ D[Phi,x], D[Phi,y] ] = 0

where the non-convex hamiltonnian is

H[X,Y]= (vx*X*X+vy*Y*Y)/sqrt[X*X+Y*Y]

In my problem vx >> vy, say vx = 100 and vy = 1

The solution of this equation (zero of Phi) is a long "ellipsoid" with
a speed of propagation on the x direction of vx=100 and in the y
direction vy=1 :

y
^
|--> x

t=0 /\
\/

t=1 ________/\_________
\/
___
t=2 ______________/ \________________
\___/

I try different methods (Lax-Friedrichs...) but these methods can't
catch the high speed in the x-direction,
instead I get a calculated speed far below vx=100 (about 60 for
instance).

-> I hope I don't bother you, but I think this is an intersting test
problem for numerical method, because none of the methods I have test
give the correct vx speed.

-> I need help if you have already meet this kind of problem.
What kind of numerical scheme should I use ? (-> It is important to me ).

Thanks,

V. PICAUD
PhD student in applied superconductivity
IGE - BELFORT
- FRANCE -

------------------------------

From: Iain Jameson <iain.jameson@dsto.defence.gov.au>
Date: Tue, 11 Apr 2000 10:07:03 +0930
Subject: A Sum of Exponentials

Is there a closed form solution to

Sum_{n=0}^{N-1} exp(i[a*n + b*n^2])

a and b are real.

Alternatively, is there a known approximation?

Thank you

Iain Jameson

------------------------------

From: Bill Hager <hager@math.ufl.edu>
Date: Fri, 14 Apr 2000 17:36:35 -0400 (EDT)
Subject: Computational Optimization and Applications Software Forum

The Computational Optimization and Applications (COAP) Software Forum
is available at the following web site:

http://www.math.ufl.edu/~coap

This web site was provides access to benchmarks for optimization
software (maintained by Hans Mittelmann), a software guide (Hans Mittelmann
and Peter Spellucci), software connected with articles published in the
COAP journal (Neculai Andrei), and test problems. Presently, the test
problem section focuses on the problems in David Gay's netlib/lp
test set. One can download the problems in MPS format (as given in
netlib/lp), LP format (using a code of Mauricio Rescende), Matlab format
(courtesy of Timothy Davis), Harwell/Boeing format (courtesy of Timothy Davis),
AMPL format (using an awk script from netlib/ampl), and LP_Solve format
(code extracted from Michel Berkelaar's LP_Solve by Hans Mittelmann). Plots of
the sparsity pattern of the matrices can be viewed (Timothy Davis).

------------------------------

From: Peter Kravanja <Peter.Kravanja@cs.kuleuven.ac.be>
Date: Mon, 10 Apr 2000 10:47:21 +0200 (CEST)
Subject: New Book, Computing the Zeros of Analytic Functions

Dear colleagues,

I'm pleased to announce that the following book has appeared:

Peter Kravanja and Marc Van Barel

COMPUTING THE ZEROS OF ANALYTIC FUNCTIONS

Lecture Notes in Mathematics, vol. 1727, Springer, 2000.
ISBN 3-540-67162-5
http://www.springer.de/cgi-bin/search_book.pl?isbn=3-540-67162-5

Summary:

Computing all the zeros of an analytic function and their respective
multiplicities, locating clusters of zeros of analytic functions,
computing zeros and poles of meromorphic functions, and solving
systems of analytic equations are problems in computational complex
analysis that lead to a rich blend of mathematics and numerical analysis.
This book treats these four problems in a unified way. It contains not
only theoretical results (based on formal orthogonal polynomials or
rational interpolation) but also numerical analysis and algorithmic
aspects, implementation heuristics, and polished software (the package
ZEAL) that is available via the CPC Program Library. Graduate students
and researchers in numerical mathematics will find this book very readable.

Key-words:

computational complex analysis, zeros of analytic functions, formal
orthogonal polynomials, rational interpolation

------------------------------

From: T. E. Simos <tsimos@mail.ariadne-t.gr>
Date: Wed, 12 Apr 2000 03:12:34 +0300
Subject: New Book Series

Dear Colleagues,

I am the Editor of a new book series of NOVA SCIENCE PUBLISHERS entitled
"Computational and Numerical Methods in Physics".

In the present series of books we will publish research books and
textbooks on the above subject. Monographs and Proceedings of
Conferences or workshops are also welcome. Special volumes on the above
subject are also welcome.

If you are intersting to publish in the new book series of Nova Science,
I would greatly appreciate to discuss with you your idea. Please send an
email to:

tsimos@mail.ariadne-t.gr

Sincerely yours

Dr. T.E. Simos
Editor of the New Series

------------------------------

From: Sergio Lipatov <lipat@dials.ccas.ru>
Date: Thu, 13 Apr 2000 10:27:08 +-400
Subject: Manuscript on Hybrid Multiscale Simulation Technology

Dear Colleagues,

I am preparing a manuscript titled "INTRODUCTION TO THE HYBRID
MULTISCALE SIMULATION TECHNOLOGY: Application to Space and Plasma Physics"
for Springer-Verlag Publisher.
This monograph address researchers and students. The first part of the book
is devoted to computational models and numerical methods. We discuss here
the hybrid codes, which include a wide spectrum of description for ions,
positrons, dust grains, atoms and electrons: particle, gyrokinetic,
guiding center, orbit-averaging approaches, photoionization and charge
exchange effects. The detailed description of all main steps of multiscale
hybrid simulation and corresponding numerical methods is presented in this
part. The second part of this book is devoted to application of simulation
technigue to study the basic plasma phenomena: wave generation by beams and
plasma clouds, particle acceleration and dissipation processes inside
collisionless shocks and current sheets. Other topics of this part are
devoted to study the global interaction of the solar wind with nonmagnetic
planets, comets, and the local interstellar medium.

This manuscript may be considered as a textbook also. I have to find the
lecturers who may use the results of plasma (and particular hybrid)
simulations in their course, and how many students attend their course.
Do You know who may be interested to use any parts from this MS for their
lectures in your University?
I shall be happy if You would like to help me.
Thanks a lot in advance.
Best regards

Alexander Lipatov

------------------------------

From: Chi-Wang Shu <shu@cfm.brown.edu>
Date: Sat, 15 Apr 2000 08:22:43 -0400 (EDT)
Subject: Journal of Scientific Computing

Journal of Scientific Computing

Recently the editorial board of this journal has been reorganized and
expanded. This journal is published quarterly by the Kluwer Academic /
Plenum Publishers. It is an international interdisciplinary forum
for the publication of papers on state-of-the-art developments in scientific
computing and its applications in science and engineering. The journal
publishes high-quality, peer-reviewed original papers, review papers and
short communications on scientific computing. The editorial board consists
of Chi-Wang Shu and Christine Bernardi as Editors, David Gottlieb,
Bertil Gustafsson, Yvon Maday and Steven Orszag as Associate Editors,
and (incomplete list) Mohamed Amara, Eytan Barouch, Marc Benard,
Jacques Blum, Claudio Canuto, Bernardo Cockburn, Knut Eckhoff, Uriel
Frisch, Daniele Funaro, Roland Glowinski, Isaac Goldhirsch, Antony Jameson,
Martin Kruskal, Pierre Sagaut, Remi Sentis, Sauro Succi, Masahisa Tabata,
Tao Tang, Roger Temam, Jean Vila and Ye Zhou as members. Inquiries and
manuscripts (ps or pdf files only with a cover letter) can be submitted
electronically to the Editors at shu@cfm.brown.edu or bernardi@ann.jussieu.fr.

------------------------------

From: Sabine Van Huffel <Sabine.VanHuffel@esat.kuleuven.ac.be>
Date: Fri, 14 Apr 2000 15:35:23 +0200 (MET DST)
Subject: Workshop on TLS and Errors-in-variables Modeling

3rd International Workshop on

TLS and ERRORS-IN-VARIABLES MODELING

August 27--29, 2001
Arenberg castle, Leuven, Belgium

Announcement and Call for Papers

This interdisciplinary workshop is a continuation of 2 previous workshops
which were held in Leuven, Belgium, August 1991 and 1996, and aims to
bring together numerical analysts, statisticians, engineers, economists,
chemists, etc. in order to discuss recent advances in Total Least Squares
(TLS) techniques and errors-in-variables modeling.

Papers are solicited for technical sessions on the following and related
topics:
TLS Concepts and Properties: structured and constrained TLS, other norms,
misfit versus latency errors, nonlinear measurement error models,
dynamic errors-in-variables, curve fitting, statistical, numerical,
robustness and optimization aspects;
TLS Algorithms: real-time, adaptive, recursive, neural, iterative algorithms,
based on SVD or related decompositions, architectures, complexity,
accuracy, regularization, convergence;
TLS Applications: array signal and image processing, model identification,
computer vision, document retrieval, spectral analysis, harmonic
retrieval, direction finding, geology, chemistry, biomedicine.

In particular, overview papers describing recent advances on any of the
above-mentioned topics are invited. Other topics related to TLS,
errors-in-variables modeling, and their applications are also welcome.

Authors are invited to submit four copies of an extended summary (2--4 pages)
to the workshop secretariat for review. Fax number and email should be
provided if possible. Overview papers will be presented as lecture, the
other ones in poster form. Authors of accepted contributions will be asked
to prepare a version for publication in the conference proceedings.

Submission of summary: February 15,2001
Notification of Acceptance: April 1, 2001
Submission of camera-ready paper: July 1, 2001

Conference committee: Sabine Van Huffel (chairperson), Bart De Moor, Yasuo
Amemiya, Leon Gleser, Gene Golub, Bjorn Ottersten, Rik Pintelon,
G. W. (Pete) Stewart and Paul Van Dooren.

Workshop secretariat: Ida Tassens <ida.tassens@esat.kuleuven.ac.be>
Dept. of Electrical Engineering, ESAT-SISTA/COSIC,
Katholieke Universiteit Leuven,
Kardinaal Mercierlaan 94, B-3001 Leuven-Heverlee, Belgium
tel : 32/16/32.17.09 fax : 32/16/32.19.70

------------------------------

From: Osman Yasar <oyasar@brockport.edu>
Date: Wed, 12 Apr 2000 10:25:39 -0400
Subject: Faculty Position at SUNY, Brockport

State University of New York, College at Brockport
Department of Computational Science

Tenure-track Assistant Professor in Computational Science,
starting in Fall 2000. Ph.D. in Computational Science, or a
closely related discipline required. Candidates should have
a strong commitment to undergraduate and graduate education
in computational science. A strong background in computational
tools, scientific visualization, computer science, and applied
mathematics is desired. Promise of continued activity in
computational research and pursuit of external funding are
required. Ability to interact with local industries is also
expected. Review of applications will begin on April 17, 2000.
Applications will continue to be accepted until the position is
filled. Send letter of application, resume, transcript of the
highest degree earned, and three letters of recommendation to

Mr. Richard D. Meade
Faculty/Staff Recruitment Office
State University of New York, College at Brockport
350 New Campus Drive, Rm 409 Allen
Brockport, NY 14420-2929

------------------------------

From: Gavin J Pringle <gavin@epcc.ed.ac.uk>
Date: Mon, 10 Apr 2000 11:55:15 +0100 (BST)
Subject: Positions at Edinburgh Parallel Computing Centre

Current Vacancies in EPCC

Edinburgh Parallel Computing Centre (EPCC) is a technology transfer
centre in the University of Edinburgh, Scotland, which promotes the
exploitation of High Performance Computing (HPC) in industry and
academia. Established in 1990, it now has around 45 staff,
first-class facilities (including a 344-processor Cray T3E), and a
client list ranging from local companies to multi-nationals and
computer vendors.

Applications Consultants (3 posts)

Currently, three vacancies exist for applications consultants at EPCC.
Duties will include the porting and optimisation of novel,
computationally intensive applications on HPC systems, and providing
technical consultancy to both industrial and academic users. Good
communication skills are important and there are opportunities to
produce training material and technical reports.

Candidates must be competent in either Fortran or C, have good
numerical skills, and should be adaptable and self-motivated. Other
useful technical skills are a good grounding in scientific research,
computer science, numerical simulation or visualisation.

More information about these posts can be found on EPCC'c WWW pages:

Vacancies: http://www.epcc.ed.ac.uk/admin/Jobs/
Home Page: http://www.epcc.ed.ac.uk/

Please quote REF: 306223

Further particulars and details of the application procedure should be
obtained from:

Personnel Office
The University of Edinburgh
9 - 16 Chambers Street
Edinburgh EH1 1HT

Tel 0131-650-2512 (24 hour answering service)

Or apply through http://www.personnel.ed.ac.uk/

Closing date: 19 May 2000

------------------------------

From: Saleet Jafri <jafri@utdallas.edu>
Date: Thu, 13 Apr 2000 10:43:40 -0500
Subject: Postdoctoral Position at University of Texas at Dallas

Position: Research Associate
Institution: The University of Texas at Dallas
Location: Dallas, TX
Deadline: 6/1/00 or when position is filled

Project Description: The cellular basis of cardiac myocyte energetics under
normal and ischemic conditions will be investigated by studying the major
components for energy production and utilization. This will be accomplished
by pursuing an interdiciplinary approach combining experiments (33%) and an
integrative modeling study (67%) directed at understanding the
interplay between Ca-handling, force generation, membrane currents and
glycolytic and mitochondrial respiration. The modeling will be done under
the supervision of Prof. Saleet Jafri and the experiments will be done under
the supervision of Prof. Dean Sherry.

Qualifications: Candidate should have a Ph.D. or M.D. with experience in
mathematical and computational modeling of biological systems and proven
interest in experimental methods.

Application: Send curriculum vitae, the names of three references, and a
description of past research experience and research interests to the
following address:

M. Saleet Jafri
Department of Mathematical Sciences EC35
The University of Texas at Dallas
Box 830688
Richardson, TX 75083

Phone: 972-883-4436
Fax: 972-883-6622
email: jafri@utdallas.edu

UT-Dallas is an EEO/AA employer.

------------------------------

From: Iain Duff <I.Duff@rl.ac.uk>
Date: Mon, 10 Apr 2000 22:21:11 +0100 (BST)
Subject: Contents, IMA Journal of Numerical Analysis

IMA JOURNAL OF NUMERICAL ANALYSIS

Contents of Volume 20, Number 2.

A. Ostermann and M. Thalhammer
Non-smooth data error estimates for linearly implicit Runge-Kutta methods.
pp 167-184

R. Sch\"atzle
On the perturbation of the zeros of complex polynomials.
pp 185-201

B. Faermann
Localization of the Aronszajn-Slobodeckij norm and application to adaptive
boundary element methods. Part I. The two-dimensional case.
pp 203-234

A. Dur\'an and J.M. Sanz-Serna
The numerical integration of relative equilibrium solutions. The nonlinear
Schr\"odinger equation
pp 235-261

C. Clavero, J.C. Jorge, F. Lisbona and G.I. Shishkin
An alternating direction scheme on a nonuniform mesh for reaction-diffusion
parabolic problems.
pp 263-280

M. Sardella
On a coupled finite element-finite volume method for convection-diffusion
problems.
pp 281-301

K. Atkinson
The planar radiosity equation and its numerical solution.
pp 303-332

------------------------------

From: Thomas Hogan <hogan@math.ohio-state.edu>
Date: Mon, 10 Apr 2000 18:20:19 -0400 (EDT)
Subject: Contents, Journal of Approximation Theory

Table of Contents: J. Approx. Theory, Volume 103, Number 2, April 2000

D. P. Dryanov and V. Vatchev
On a polynomial inequality of Paul Erd\H{o}s
183--205

Pradipta Bandyopadhyay and T. S. S. R. K. Rao
Central subspaces of Banach spaces
206--222

Grzegorz W. Wasilkowski and Henryk Wo\'zniakowski
Complexity of weighted approximation over $\RR^1$
223--251

D. L. Ragozin and J. Levesley
The density of translates of zonal kernels on compact homogeneous
spaces
252--268

Michael Revers
The divergence of Lagrange interpolation for $|x|^\alpha$ at
equidistant nodes
269--280

Ying Guang Shi
On Christoffel type functions for $L_m$ extremal polynomials, I
281--291

Salvador Romaguera and Manuel Sanchis
Semi-Lipschitz functions and best approximation in quasi-metric spaces
292--301

Ingo Steinwart
Entropy of $C(K)$-valued operators
302--328

Dietrich Braess and Christoph Schwab
Approximation on simplices with respect to weighted Sobolev norms
329--337

Peter G. Casazza and Ole Christensen
Approximation of the inverse frame operator and applications to
Gabor frames
338--356

Note

J. Szabados
Where are the nodes of good'' interpolation polynomials on the
real line?
357--359

Author index for Volume 103
360

------------------------------

From: Hans Schneider <hans@math.wisc.edu>
Date: Wed, 12 Apr 2000 19:49:46 -0500 (CDT)
Subject: Contents, Linear Algebra and Its Applications

ContentsDirect from Elsevier Science

URL: http://www.elsevier.nl/locate/jnlnr/07738

Journal: Linear Algebra and Its Applications
ISSN : 0024-3795
Volume : 308
Issue : 1-3
Date : 13-Apr-2000

pp 1-29
Normal forms of generic triangular band matrices and Jordan forms of
nilpotent completions
MI Gekthman, L Rodman

pp 31-64
Generalized-confluent Cauchy and Cauchy-Vandermonde matrices
ZHENGHONG Yang

pp 65-75
Pascal k-eliminated functional matrix and it's property
M Bayat

pp 77-84
Matrix young inequalities for the Hilbert-Schmidt norm
O Hirzallah, F Kittaneh

pp 85-107
Representation and approximation of the outer inverse A_T,S^(2) of a
matrix A
YONGLIN Chen

pp 109-119
Metric projection and stratification of the Grassmannian
M Finzel

pp 121-137
Spectrally arbitrary patterns
JH Drew, P Van Den Driessche

pp 139-152
Decomposable numerical ranges on orthonormal tensors
CK Li

pp 153-161
Estimates for the spectrum near algebraic elements
Y Chen, T Ransford

pp 163-181
An extension problem for discrete- time almost periodically
correlated stochastic processes
D Alpay

pp 183-202
Pade approximation for the exponential of a block triangular matrix
L Diece

pp 203-211
Notes on matrix arithmetic- geometric mean inequalities
R Bhatia

pp 217-217
Author index

Journal: Linear Algebra and Its Applications
ISSN : 0024-3795
Volume : 309
Issue : 1-3
Date : 20-Apr-2000

Proceedings of the international worskhop on accurate solutions of eigenvalue
problems
Special editors: Jesse L. Barlow, Beresford N. Parlett, Kresimir Veslic

pp 1-2
Preface: The Accurate solution of Eigenvalue Problems
JL Barlow

pp 3-18
Weyl-type relative perturbation bounds for eigenvalues of Hermitian
matrices
FM Dopico, JM Molera

pp 19-43
Optimal perturbation bounds for the Hermitian eigenvalue problem
JL Barlow

pp 45-56
Absolute and relative perturbation bounds for invariant subspaces of
matrices
ICF Ipsen

pp 57-72
Relative perturbation theory for hyperbolic eigenvalue problem
I Slapnicar

pp 73-83
Accuracy assessment for eigencomputations: variety of backward errors
and pseudospectra
F Chaitin-Chatelin, E Traviesas

pp 85-102
Perturbation theory for the eigenvalues of factorised symmetric
matrices
K Veselic

pp 103-119
Rounding-error and perturbation bounds for the indefinite QR
factorization
S Singer

pp 121-151
Relatively robust representations of symmetric tridiagonals
BN Parlett

pp 153-174
QR factorization with complete pivoting and accurate computation of
the SVD
NJ Higham

pp 175-190
Exponential splittings of products of matrices and accurately
computing singular values of long products
S Oliveira, DE Stewart

pp 191-215
Approximate eigenvectors as preconditioner
Z Drmac

pp 217-259
An implementation of dqds algorithm (positive case)
BN Parlett

pp 261-287
Balancing sparse matrices for computing eigenvalues
TY Chen, J Demmel

pp 289-306
Large sparse symmetric eigenvalue problems with homogeneous linear
constraints: the Lanczos process with inner-outer iterations
GH Golub, H Zha

pp 307-323
The influence of orthogonality on the Arnoldi method
T Braconnier

pp 325-337
The relative error in the Pruess method for Sturm-Liouville problems
P Kosowski

pp 339-361
Backward error and condition of polynomial eigenvalue problems
F Tisseur

pp 363-363
Author index

NOTE:
ContentsDirect, which is automatically generated, lists the first author
of each paper and the corresponding author (if different).

------------------------------

End of NA Digest

**************************
-------