NA Digest Sunday, August 4, 1996 Volume 96 : Issue 29

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: NA Digest <>
Date: Sun Aug 4 13:58:43 EDT 1996
Subject: NA Digest Calendar

The Netlib Conferences Database is on the Web at:

NA Digest Calendar
Date Topic Place NA Digest #

Aug. 1- 3 Structured Matrices Santa Barbara, CA 22
Aug. 5- 7 Computational Fluid Dynamics Hampton, VA 17
Aug. 5-13 Industrial Mathematics Modeling Raleigh, NC 13
Aug. 7- 8 Brazil Interval Workshop Recife, Brazil 18
Aug. 14-17 International Linear Algebra Society Chemnitz, Germany 15
Aug. 18-21 Applied Parallel Computing Lyngby, Denmark 24
Aug. 19-21 High Performance Computing Ithaca, NY 21
Aug. 19-21 Parallel Irregular Problems Santa Barbara, CA 44
Aug. 21-24 Total Least Squares Leuven, Belgium 25
Aug. 23 Computational Finance Conference Palo Alto, CA 02
Aug. 25-31 Congress Theor. & Appl. Mechanics Kyoto, Japan 94:46
Aug. 26-30 Lectures in Computational Fluid Dynamics Montreal, Canada 25
Aug. 27-29 Parallel Numerical Algorithms Lyon, France 04
Aug. 27-29 Programming Environment and Tools Lyon, France 06

Sep. 1- 5 Numerical Solution of ODEs Halle, Germany 13
Sep. 2- 5 Nonlinear Programming Beijing, China 04
Sep. 4- 6 Multidisciplinary Analysis Bellevue, WA 05
Sep. 4- 6 Operations Research Braunschweig, Germany 26
Sep. 8-11 Computing in Europe on IBM Platforms Krakow, Poland 14
Sep. 9-13 "Diskrete Optimierung" Hamburg, Germany 18
Sep. 9-13 Computational Methods Applied Sciences Paris, France 15
Sep. 9-14 Inverse and Ill-Posed Problems Moscow, Russia 01
Sep. 9-27 Numerical Simulation of PDEs Trieste, Italy 04
Sep. 15-17 Control System Design Dearborn, MI 43
Sep. 16-18 Multivalued Eikonal Solvers Rocquencourt, France 06
Sep. 16-18 Modern Software Tools Oslo, Norway 24
Sep. 16-21 Computational Modelling Dubna, Russia 07
Sep. 17-20 Italian Operational Research Society Perguia, Italy 05
Sep. 18-20 Virtual Systems and Multimedia Gifu, Japan 14
Sep. 19-29 High Performance Optimization Techniques Delft, Netherlands 22
Sep. 21 Pacific Northwest NA Seminar Vancouver, B. Columbia 26
Sep. 23 Computational Mathematics Glasgow, Scotland 18
Sep. 23-27 Inverse Problems in Wave Propagation Aix les Bains, France 03
Sep. 24-26 Boundary Element Method Braga, Portugal 03
Sep. 24-27 Modeling Issues for Environment Albuquerque, NM 07
Sep. 25-27 Vector and Parallel Processing Porto, Portugal 51
Sep. 25-27 Dutch Numerical Mathematics Zeist, Netherlands 19
Sep. 25-28 Mathematical Tools in Metrology Berlin, Germany 25
Sep. 26-27 Air Pollution in Europe Copenhagen, Denmark 28
Sep. 26-28 Computer Mathematics Athens, Greece 09
Sep. 29... IBM Parallel Programming Ithaca, NY 25
Sep. 30... Interval Methods Wuerzburg, Germany 13

Oct. 1- 4 European Multigrid Conference Stuttgart, Germany 20
Oct. 3- 4 Parallel Computing Minneapolis, MN 29
Oct. 7- 9 ICASE/LaRC Industry Roundtable Williamsburg, VA 27
Oct. 9-11 Workshop on Scientific Computing Braunschweig, Germany 24
Oct. 9-11 SIAM Conference on Sparse Matrices Coeur d'Alene, ID 16
Oct. 10-11 International Meshing Roundtable Pittsburgh, PA 15
Oct. 20-23 High Performance Computing Tempe, AZ 01
Oct. 21-22 Computational Science and Engineering Purdue, IN 24
Oct. 21-25 Evolutionary Algorithms Minneapolis, MN 29
Oct. 24-26 Materials Studies Workshop University Park, PA 27

Nov. 6- 8 Innovative Time Integrators Amsterdam, Netherlands 40
Nov. 18-21 Overset Grids Symposium Los Alamos, NM 27

Dec. 17-19 Mathematics in Signal Processing Warwick, England 48


Jan. 5- 7 Discrete Algorithms New Orleans, LA 15
Jan. 5-12 Computational Mathematics Rio de Janeiro, Brazil 08
Jan. 5-12 Numerical Linear Algebra Rio de Janeiro, Brazil 13
Jan. 24-26 Multi-Scale Problems Kiel, Germany 21
Jan. 27-31 Maths-in-Industy Study Group Melbourne, Australia 11
Mar. 10-12 Scientific Computing Hong Kong 25
Mar. 12-14 Algorithms and Complexity Rome, Italy 24
Mar. 14-17 SIAM Parallel Processing Minneapolis, MN 29
Mar. 16-21 Approximation and Optimization Caracas, Venezuela 21
Mar. 21-22 AMS Session on Approximation Theory Memphis, TN 11
Apr. 1- 3 Monte Carlo Methods Brussels, Belgium 16
Apr. 6-11 Multigrid Copper Mountain, CO 19
Apr. 17-18 Meeting Honoring Bill Morton Oxford, England 26
May 19-21 Applications of Dynamical Systems Snowbird, UT 27
May 26-30 Computational Heat Transfer Cesme, Turkey 05
June 16-18 Mathematical Issues in Geosciences Albuquerque, NM 18
June 18-21 Principles + Practice of Parallel Prog. Las Vegas, NV 27
June 24-27 Dundee NA Conference Dundee, Scotland 13
July 14-18 Theoretical and Computational Acoustics New York, NY 14
Aug. 24-29 IMACS World Congress Berlin, Germany 07
Sep. 15-18 Boundary Integral Methods Manchester, England 27
Sep. 15-19 Scientific Computing & Diff. Eqns. Grado, Italy 26
Sep. 29-.. ENUMATH-97 Heidelberg, Germany 50


From: Nick Trefethen <lnt@CS.Cornell.EDU>
Date: Thu, 1 Aug 1996 10:04:35 -0400
Subject: Trefethen ODE/PDE Textbook

Recently I have received many inquiries concerning my unpublished
textbook "Finite Difference and Spectral Methods for Ordinary and
Partial Differential Equations," the basis of graduate courses
taught at MIT and Cornell and a few other places since 1985.
Here's the status: the book has not been finished, and I do not
expect to finish it in the next few years. However, the portion
that exists (325 pages, with numerous exercises) is in reasonably
polished form and suitable for classroom use.

Tired of xeroxing copies on an individual basis, I have decided to
have this unfinished book printed and bound for general availability.
To obtain a copy, send me 22 US dollars or 15 pounds sterling (this
includes shipping charges). A table of contents and excerpts from
the book are available from my homepage (URL below). Not all of the
book is in PostScript, so it would not be possible to make it fully
available on the Web.

Nick Trefethen
Cornell University


From: Robert van de Geijn <>
Date: Fri, 2 Aug 1996 12:14:12 -0500
Subject: Parallel Linear Algebra Package, PLAPACK

We are approaching the beta release of our new package, Parallel
Linear Algebra Package (PLAPACK). The primary differences between
this library and earlier libraries include the adoption of a more
application friendly data distribution (Physically Based Matrix
Distribution) and a programming style that avoids the use of indices.
We believe this approach makes implementation of parallel dense linear
algebra libraries in some sense simpler than MATLAB script language.

While the beta release will likely not happen until early September,
we believe the Users' Guide may be of some use to those who teach
practical courses on the subject this fall. For this reason, we make
the manuscript in preparation available via the PLAPACK webpage:

By filling out the guestbook, one can request to become a beta site.

Robert van de Geijn

Robert A. van de Geijn
Associate Professor
Department of Computer Sciences (Work) (512) 471-9720
The University of Texas (Home) (512) 251-8301
Austin, TX 78712 (FAX) (512) 471-8885


From: Steve Sullivan <>
Date: Mon, 29 Jul 1996 15:51:51 -0600
Subject: Numerical Analysis Frequently Asked Questions

FAQ on Numerical Analysis & Associated Fields Resource Guide

Nick Mikhailovsky has made an HTML version of the NA FAQ,
which is available at:

It is an evolving document, and we welcome suggestions.
A partial list of contents is:

Indices of Software
Libraries on the Net
Packages on the Net
Commercial Libraries and Packages
Professional Societies and Their Publications
Electronic Newsletters
Electronic Journals
Online Preprints
Miscellaneous Web Sites for NA
Books, With and Without Software

Linear Algebra
Random Number Generators (RNGs)
Function evaluation
Finding Roots
Optimization, minimization
Curve Fitting, Data Modelling, Interpolation, Extrapolation
Transforms (FFT, etc) and digital signal processing (DSP)
Integration and Ordinary Differential Equations (ODEs)
N-Body and Particle Simulation
Partial Differential Equations (PDEs) and Finite Element Modeling (FEM)
Operations Research: minimization, optimization
Graphics and Scientific Visualization
Miscellaneous NA Software

Probability and Statistics
Chaos Theory (Nonlinear Dynamics)
Symbolic Algebra
Other Associated Fields

Steve Sullivan


From: Greg Landsberg <>
Date: Mon, 29 Jul 1996 12:18:41 -0500 (CDT)
Subject: Printed Circuit and Semiconductor Simulation Code Needed

Hi Folks,

I need to solve a 3-dimensional Laplace equation on the domain

0 < x < x_0 (x_0 << y_0)
0 < y < y_0
-\infty < z < +\infty,

with the following boundary conditions: periodicity in z with alternating
Neumann and Dirichlet conditions on the x=0 plane. I have a mixture of
Neumann and Dirichlet conditions along x- and y-directions.

This problem originates from calculating electrostatic parameters of a long
(z) thin (x) cable made of narrow (y) metal strips glued to the dielectric.
The x=0 plane is made of a metal mesh, which gives rise to periodicity in
boundary conditions on x=0 plane and along the z-direction.

Could anyone point me toward available programs for solving this problem
with <5% accuracy. This should be a standard problem of designing printed
circuits or large semiconductor devices.

Thanks so much in advance!

Greg Landsberg
Fermi National Accelerator Laboratory
MS #357
Batavia, IL 60540


From: Workshop HPOP <>
Date: Mon, 29 Jul 1996 12:09:41 METDST
Subject: High Performance Optimization Techniques

Second Announcement

Stieltjes workshop on
High Performance Optimization Techniques
Thursday, September 19 to Friday, September 20, 1996

To be held at
The Delft University of Technology,
Delft, The Netherlands

This second announcement contains more information about main speakers
and already submitted papers, plus registration information.

Aim of the workshop

Recent progress in computer technology on the one hand and algorithms
or optimization on the other hand has led to a dramatic increase of the
efficiency of solution methods for a wide class of optimization
problems. The aim of the workshop is to bring together some of the most
active researchers working on the design and implementation of
optimization algorithms.
We aim to cover the latest algorithmic developments, complexity results
and implementation aspects, including the required tools from numerical
Much attention will be given to the recent developments in semidefinite
optimization and its relevance for a wide range of practical
applications in fields such as combinatorial optimization, engineering,
design, matrix inequalities in systems and control theory, and matrix
completion problems.

Main speakers

M.X. Goemans (M.I.T.)
Semidefinite Programming in Combinatorial Optimization.
F. Rendl (Technische Universitat Graz)
Solving integer programs using SDP.
Ph. Toint (Universite de Namur)
1: A view of the state of the art in constrained nonlinear
2: A new primal-dual-method for nonconvex problems with bounds and
linear equality constraints.
L. Vandenberghe (Stanford University)
Semidefinite programming: engineering applications and primal-dual
interior-point algorithms.

More papers already submitted

Kurt Anstreicher (University of Iowa)
Continuous relaxations for constrained maximum-entropy sampling.
Arjan B. Berkelaar (Erasmus University Rotterdam)
Polynomial Cone Affine Scaling Algorithm for Semidefinite
E. de Klerk (TU Delft)
Polynomial primal-dual affine scaling algorithms in semidefinite
Monique Laurent (CWI/CNRS)
Positive semidefinite and Euclidean distance matrix completion
Ivan Raykov (Sheffield Hallam Univ./Techn. Univ. of Varna)
A Parametric Newton Method for Solving Unconstrained Nonlinear
Optimizations Problem in Hilbert Space.
Katya Scheinberg (Columbia University)
On Parametric Semidefinite Programming
Jos F. Sturm (Erasmus University Rotterdam)
On the long step path-following method for semidefinite programming.
T. Terlaky (T.U. Delft)
Embedding of SD.
Dr. S. Zhang (Erasmus University)
On weighted centers for semidefinite programming.

Program Committee

Jan Karel Lenstra (Technische Universiteit Eindhoven)
Kees Roos (Delft University of Technology)
Lex Schrijver (CWI)
Tamas Terlaky (Delft University of Technology)
Henk A. van der Vorst (Rijksuniversiteit Utrecht)

Registration and further Information

Registration for the workshop is open until August 9, 1996. The
registration fee is Dfl 150, including lunches. The easiest way to
register is the electronic registration form at our web-site. You may
also ask us for an e-mail registration form. The number of participants
is restricted to 50.
If you wish to give a presentation, please provide a title with your
registration. Abstracts of talks should be submitted before August 30,

More information about the workshop, including a tentative list of
participants and presentations, and also some travelling and hotel
information can be found at our web-site

For any questions or remarks, please contact us via e-mail or at the
address below.

Address information

Organizing Committee HPOPT
Delft University of Technology
Faculty of Mathematics
Department of Operations Research
Mekelweg 4
2628 CD Delft, The Netherlands

tel. +31 15 278 2547
fax +31 15 278 7255

Local Organizing Committee

Etienne de Klerk
Arie Quist
Kees Roos
Tamas Terlaky
Netty Zuidervaart-Murray


From: Trini Flores <>
Date: Wed, 31 Jul 96 14:38:19 EST
Subject: SIAM Conference on Parallel Processing

Eighth SIAM Conference on Parallel Processing for Scientific Computing
March 14-17, 1997
Hyatt Regency Minneapolis on Nicollet Mall
Minneapolis, Minnesota

Sponsored by SIAM Activity Group on Supercomputing

Co-chairs: Michael Heath, University of Illinois, Urbana
Virginia Torczon, College of William and Mary

We invite you to participate in the conference. Submit your abstracts
now to


September 16, 1996: Deadline for submission of minisymposium

October 1, 1996: Deadline for submission of contributed abstracts.

For additional information contact: SIAM, telephone: 215-382-9800;
e-mail: WWW:


From: Michael Olesen <>
Date: Thu, 1 Aug 1996 09:51:31 -0500 (CDT)
Subject: International Conference on Parallel Computing

Preliminary announcement

October 3-4, 1996
University of Minnesota Supercomputer Institute
1200 Washington Avenue South
Minneapolis, Minnesota

sponsored by:
Supercomputer Institute and
IBM Corporation

The International Conference on Parallel Computing will bring
together scientists to discuss distributed parallel applications in
the computational sciences and to foster cross-disciplinary
interactions in the general area of high-performance computing.
There will be a mix of talks dealing with novel programming and
algorithmic aspects of parallel computing as well as state-of-the-
art technical advances. An international group of invited speakers
is currently being contacted. The conference will provide a
concentrated forum to discuss progress on enhancing the number of
applications that can take advantage of emerging parallel computing

Confirmed Speakers (to date):

Robert Beck, University of Chicago
"An Overview of Parallel Processing in Biomedical Imaging"
Gyan Bhanot, IBM Yorktown
"A Cross Section of 'Super' Research done on Computers at the
IBM TJ Watson Research Center"
Jeremy Broughton, Naval Research Laboratory
"Direct Atomistic Simulation of Next Generation Quartz
Crystal Oscillators"
David Ceperley, University of Illinois
"Path Integral Monte Carlo Calculations with Message Passing"
Michel Dupuis, Pacific Northwest National Laboratory
"Advanced Techniques For High Scalability: Application to
Computational Chemistry"
Dennis Evans, Australian National University
"Non-Equilibrium Molecular Dynamics Calculation of Transport
Coefficients in Fluids
Alan Needleman, Brown University
"Computational Modeling of Material Failure"
Shuhei Ohnishi, NEC Fundamental Research Laboratories, Tsukuba,
"Materials Construction by the Density Functional Theory with
the Aid of Parallel Processors"
Yousef Saad, University of Minnesota
"Sparse Iterative Methods on Distributed Memory Computers"
Ulrich Trottenberg, German National Research Center for
Information Technology, St. Augustin, Germany
"Parallelism and Adaptivity -- Contradiction?"
Priya Vashishta, Louisiana State University
"Large-Scale Simulation of Crack Propagation and Fracture on
Parallel Computers"

There is no conference fee but participants must register to attend
because of the space limitation and lunch arrangements. A $45 fee
will be charged for the Thursday night, October 3, banquet. Up-to-
date information may be found at For other information
or to register please contact the symposium administrator, Michael
Olesen, 612-624-1356.

Preregistration is required.

Michael J. Olesen
Research Programs Administrator
University of Minnesota
Supercomputer Institute
1200 Washington Avenue South
Minneapolis, MN 55415

office: (612) 624-1356
fax: (612) 624-9565


From: Michael Olesen <>
Date: Thu, 1 Aug 1996 11:25:21 -0500 (CDT)
Subject: Workshop on Evolutionary Algorithms


Dates: October 21-25, 1996

Conference Site: Institute for Mathematics and its Applications
University of Minnesota
Minneapolis, Minnesota

Sponsors: Institute for Mathematics and its Applications
Supercomputer Institute


The workshop will bring together key researchers for a concentrated
discussion of theory and applications of evolutionary algorithms,
including genetic algorithms, evolution strategies, and evolutionary
programming. The workshop will be organized around a limited number of
plenary lectures with generous blocks of time for interactive discussion

The workshop will begin with a series of expository talks that overview
the foundations of these fields and define critical issues to be
discussed during the workshop. A major goal is to use discussion time to
communicate recent research results and to bridge the gap between theory
and applications. The workshop will also assess the state of the current
mathematical underpinning of the field and reconcile some of the diverse
theoretical perspectives of the various participants. The workshop will
emphasize high-performance computing using parallel evolutionary
algorithms and hybrid algorithms in critical state-of-the-art applications.

The workshop will be of interest to:

* practitioners from industry with applications in
search, optimization, and machine learning
* mathematicians working in dynamical systems and Markov processes
* people interested in application areas and algorithms for
high-performance computing

Up-to-date information about this conference is available at

For information on how to register, contact the IMA at:

Institute for Mathematics and its Applications
514 Vincent Hall
206 Church Street SE
Minneapolis, MN 55455

Michael J. Olesen
Research Programs Administrator
University of Minnesota
Supercomputer Institute
1200 Washington Avenue South
Minneapolis, MN 55415

office: (612) 624-1356
fax: (612) 624-9565


From: Gerald Browning <>
Date: Mon, 29 Jul 1996 10:02:30 -0600
Subject: Position Available at NOAA in Boulder, Colorado

Applied Mathematical Research Position in Boulder, CO

Position Description:

A scientific research position is available for one year
with the possibility of extension to at least two additional
years. The position involves the development of a composite
mesh ocean model based on the new system of equations
developed by Browning and Kreiss (Dynamics of Atmospheres
and Oceans, 1990, 14, 303-332). Applicants should possess a applied mathematics or numerical analysis with some
training in the theory and numerical approximation of partial
differential equations.The successful candidate will also
have experience in solving numerical approximations
of partial differential equations using Fortan and/or C++.

Appointments will be through the Cooperative Institute
for Research in the Atmosphere at Colorado State University,
but the research will be performed at NOAA in
Boulder Colorado. The salary will depend on the applicants
qualifications (salary range is $30,00 - $45,000).

Position is available October 1, 1996.
To apply send a detailed CV and arrange to have at least
two letters of reference sent to

Gerald Browning
325 Broadway
Boulder, CO 80303


From: SIAM <>
Date: Mon, 29 Jul 96 08:32:19 EST
Subject: Contents, SIAM Applied Mathematics

Applied Mathematics
OCTOBER 1996 Volume 56, Number 5

Metastable Bubble Solutions for the Allen-Cahn Equation with Mass Conservation
Michael J. Ward

Diffusion Limits for Flows in Thin Layers
Hans Babovsky

Unsteady Gasdynamic Evolution of an Induction Domain Between a Contact Surface
and a Shock Wave. I: Thermal Runaway
Mark Short and J. William Dold

The Onset of Thermal Convection Between Poorly Conducting Horizontal Boundaries
in the Presence of a Shear Flow
Stephen M. Cox

The Linearization of the Dirichlet-to-Neumann Map in the Anisotropic
Kirchhoff-Love Plate Theory
Masaru Ikehata

Perturbing the Critically Damped Wave Equation
Steven J. Cox and Michael L. Overton

The Moving Singularities of the Perturbation Expansion of the Classical Kepler
Mohammad Tajdari

Lifted Lattices, Hyperbolic Structures, and Topological Disorders in Coupled Map
Wenxian Shen

Moment Lyapunov Exponent for Two Coupled Oscillators Driven by Real Noise
N. Sri Namachchivaya, H. J. Van Roessel, and M. M. Doyle

Statistical Inverse Estimation in Hilbert Scales
Bernard A. Mair and Frits H. Ruymgaart

Parabolic and Gaussian White Noise Approximation for Wave Propagation in Random
F. Bailly, J. F. Clouet, and J. P. Fouque

Exact and Asymptotic Solutions for the Time-Dependent Problem of Collective Ruin
Charles Knessl and Craig Steven Peters


From: Ron Boisvert <>
Date: Thu, 1 Aug 96 12:00:58 EDT
Subject: Contents, Transactions on Mathematical Software

For details, see the TOMS Web page at

Table of Contents
Volume 22, Number 3 (September 1996)

T. E. Hull and R. Mathon
The Mathematical Basis and a Prototype Implementation of a New
Polynomial Rootfinder with Quadratic Convergence

Maria Sosonkina, Layne T. Watson and David E. Stewart
Note on the End Game in Homotopy Zero Curve Tracking

Allan J. MacLeod
Algorithm 757: MISCFUN, A Software Package to Compute Uncommon
Special Functions

J. G. Blom and R. A. Trompert and J. G. Verwer
Algorithm 758: VLUGR2, A Vectorizable Adaptive Grid Solver for PDEs
in 2D

J. G. Blom and R. A. Trompert and J. G. Verwer
Algorithm 759: VLUGR3, A Vectorizable Adaptive Grid Solver for PDEs
in 3D

Knud D. Andersen
A Modified Schur-Complement Method for Handling Dense Columns in
Interior-Point Methods for Linear Programming

Hiroshi Akima
Algorithm 760: Rectangular-Grid-Data Surface Fitting that has the
Accuracy of a Bicubic Polynomial

Hiroshi Akima
Algorithm 761: Scattered-Data-Fitting that has the Accuracy of a
Cubic Polynomial

Barry W. Brown, Floyd M. Spears, Lawrence B. Levey, James Lovato and
Kathy Russell
Algorithm 762: LLDRLF, Log-Liklihood and Some Derivatives for Log-F


End of NA Digest