NA Digest Saturday, May 11, 2002 Volume 02 : Issue 19

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

Submissions for NA Digest:

Mail to

Information via e-mail about NA-NET: Mail to


From: Arnold Neumaier <>
Date: Fri, 10 May 2002 12:26:58 +0200
Subject: The Mathematics Genealogy Project

I came across a very interesting WWW-site:

The Mathematics Genealogy Project

The site contains names, advisors, and titles of dissertations
of mathematicians with a Ph.D., throughout the history

Currently (May 6th, 2002) there are records of 55249 mathematicians.
While fairly complete on US mathematicians, the directory
is still quite incomplete for the rest of the world.
The authors invite submissions updating incomplete or missing

Please contribute and make the directory more complete and
hence more useful.

Arnold Neumaier


From: Carlos deMoura <>
Date: Mon, 06 May 2002 22:16:52 -0300
Subject: Change of Address for Carlos A. de Moura

Dear colleagues,

Thank you for updating your files with my new addresses:

Carlos A. de Moura
Institute for Mathematics and Statistics - IME
Rio de Janeiro State University - UERJ
Phone: 5521 25877460
Av. S. Francisco Xavier 524
22550-900 Maracana, Rio de Janeiro


From: David Grubbs <>
Date: Mon, 6 May 2002 15:39:00 -0400
Subject: New Book, Mathematical Theory of Finite Element Methods

Springer-Verlag is proud to announce the April publication of The
Mathematical Theory of Finite Element Methods, Second Edition by Susanne
C. Brenner and L. Ridgway Scott in our Texts in Mathematics series.

This book develops the basic mathematical theory of the finite element
method, the most widely used technique for engineering design and
analysis. This expanded second edition contains new chapters on additive
Schwarz preconditioners and adaptive meshes. New exercises have also
been added throughout.

The book will be useful to mathematicians as well as engineers and
physical scientists. It can be used for a course that provides an
introduction to basic functional analysis, approximation theory, and
numerical analysis, while building upon and applying basic techniques of
real variable theory.

The initial chapter provides an introduction to the entire subject,
developed in the one-dimensional case. Four subsequent chapters develop
the basic theory in the multidimensional case, and a fifth chapter
presents basic applications of this theory.

Subsequent chapters provide an introduction to:
multigrid methods
domain decomposition methods
mixed methods with applications to elasticity and fluid flow models
iterated penalty method and augmented Lagrangian method
variational "crimes" including nonconforming and isoparametric methods
and numerical integration
error estimates in the maximum norm with applications to nonlinear problems
error estimators and adaptive meshes
Banach-space operator-interpolation techniques

Different course paths can be chosen, allowing the book to be used for
courses designed for students with different interests. For example, courses
can emphasize physical applications, or algorithmic efficiency and code
development issues, or the more difficult convergence theorems of the subject.


From: Joerg Steinbach <>
Date: Tue, 7 May 2002 16:37:12 +0200
Subject: New Book, A Variational Inequality Approach to Free Boundary Problems

I am pleased to announce a new book in International Series of Numerical
Mathematics (ISNM 136), titled

A Variational Inequality Approach to Free Boundary
Problems with Applications in Mould Filling

by Joerg Steinbach.

Published in April 2002 by Birkhaeuser Verlag AG, Switzerland.

The monograph is devoted to the study of an evolutionary variational
inequality approach to a degenerate moving free boundary problem both with
respect to analytical and numerical methods. The inequality approach of
obstacle type is the result of the application of an integral transformation
to the free boundary problem. Both elliptic and parabolic variational
inequalities are well studied in literature, but the presented type, which is
especially characterized by the memory term, takes an intermediate position
between the types just mentioned. Therefore, a special treatment is required.
The study of such inequality problems is motivated by their applications,
e.g., a generalized Hele-Shaw flow in injection and compression moulding,
the electro-chemical machining process with a time-dependent conductivity
or a quasi-stationary Stefan type problem with zero-specific heat.

The mathematical analysis of the considered problem comprises existence,
uniqueness, regularity and the time evolution of the solution in the
framework of the variational inequality theory.

Both finite element and finite volume approximations are analyzed in a
variational framework for the numerical solution of the evolutionary inequality
problem in two and three space dimensions. This includes also a comparison of
these approximations for elliptic as well for the evolutionary obstacle problem.
Results of numerical experiments are presented to illustrate the convergence

Finally, there is given an overview on the mathematical modelling of injection
and compression moulding by means of a generalized Hele-Shaw flow. Moreover,
features of different mathematical models (distance concept, Navier-Stokes flow)
are discussed and compared, where especially the (industrial) application
point of view is emphasized.

Check the link: for more

Please send comments to my new e-mail:


From: Brynjulf Owren <>
Date: Tue, 07 May 2002 13:28:25 +0200
Subject: SciCADE Conference 2003 in Trondheim

Conference on Scientific Computing and Differential Equations SciCADE'03
The Norwegian Institute of Science and Technology, Trondheim, Norway,
June 30 - July 4, 2003.

This meeting is concerned with scientific computing involving practical and
theoretical aspects of the numerical solution of differential equations:

* ordinary differential equations
* partial differential equations
* dynamical systems
* differential algebraic equations
* software
* mathematical structures related to numerical methods

Numerical techniques in applications will be emphasized. These include:

o optimization and optimal control
o inverse problems
o chemical and mechanical engineering
o stochastic differential equations
o systems with structure preservation
o molecular dynamics
o computer graphics
and more.

Previous Scicade meetings have been held approximately every two years,
with the three most recent taking place in Grado (Italy), Fraser Island
(Queensland), and Vancouver (Canada)

The meeting format includes plenary talks, minisymposia and contributed
talks in the SIAM or ICIAM pattern. Confirmed plenary speakers are

Wolf-Juergen Beyn, University of Bielefeld, Germany
Christian Brouder, IMCP, Paris, France
Bjorn Engquist, UCLA, USA
Omar Ghattas, Carnegie Mellon University, USA
Arieh Iserles, Cambridge University, UK
Ben Leimkuhler, University of Leicester, UK
Bob Russel, Simon Fraser University, Canada
Volker Schulz, University of Trier, Germany
Andrew Stuart, University of Warwick, UK
Antoinette Tordesillas, University of Melbourne, Australia
Gerhard Wanner, Universite de Geneve, Switzerland

The Dahlquist prize will be presented by SIAM.

Please visit our web site for more details at

or drop us a line at

We very much hope you will take part in this event.


From: Stefano Micheletti <>
Date: Tue, 7 May 2002 10:25:13 +0200
Subject: Workshop in Milano on Anisotropic Grids

We have the pleasure to announce the one-day workshop:

"Anisotropic Grids: Generation, Adaption and Error Estimation".
Organizers: Carlo L. Bottasso, Stefano Micheletti, Simona Perotto,
Riccardo Sacco

Friday, June 21, 2002, 9:30 a.m. to 5:00 p.m.
MOX (Modeling and Scientific Computing), Palazzo "La Nave", 6th floor,
Dipartimento di Matematica, Politecnico di Milano, via Bonardi 9, 20133
Milano, Italy.

Topics of the workshop include:
- mesh generation algorithms for anisotropic problems (e.g., boundary layer
- mesh adaption algorithms for tracking anisotropic solution features, such
as shock waves, wakes, etc.;
- error estimation procedures that can effectively drive anisotropic mesh
adaption schemes;
- applications and industrial problems involving anisotropic solution

The meeting aims at bringing together a limited number of participants that
share a common interest on these topics. The informal nature of the workshop
will provide an opportunity for close interaction and profitable discussion.
The format of the meeting will include seven 30-minute presentations
and a final session in the afternoon open for discussion on problems,
new trends, recent ideas, etc. This final session should be the occasion
for questions and discussions concerning the contents of each talk and
should hopefully allow for a more ample interaction among
the participants than normally possible in larger conferences.

The list of speakers include:

C.L. Bottasso (Politecnico di Milano)
L. Formaggia (EPFL)
P. Frey (INRIA)
G. Kunert (TU Chemnitz)
S. Micheletti (Politecnico di Milano)
J. Peiro' (Imperial College)
V. Selmin (Alenia Aerospazio)

For further information please visit the MOX web site:

The registration fee is 50 Euros.
To register, you are kindly invited to contact:

Pamela Palazzini (MOX Secretary),
Dipartimento di Matematica, "F. Brioschi", Politecnico di Milano, Italy.
Tel.: +39-02-2399-4611

We look forward to meeting you in Milano.


From: Ken Jackson <>
Date: Tue, 7 May 2002 09:39:10 -0400
Subject: Workshop in Toronto on PDEs on the Sphere

Second Announcement and Call for Papers
The 2002 Workshop on
The Solution of Partial Differential Equations on the Sphere
August 12-15, 2002
The Fields Institute, Toronto

We would like to remind you that the deadline for abstracts for the
2002 Workshop on The Solution of Partial Differential Equations on the
Sphere is May 15.

The workshop will be held August 12-15, 2002, hosted by The Fields
Institute for Research in Mathematical Sciences, Toronto, Canada, as
part of the Thematic Year on Numerical and Computational Challenges in
Science and Engineering.

Information on the workshop and the Thematic Year in general can be
found at:

We hope to see you in Toronto in August 2002.

David Williamson (NCAR)
Ken Jackson (University of Toronto)


From: David Dureisseix <>
Date: Fri, 10 May 2002 09:51:01 +0200
Subject: Workshop in Cachan on Multiscale Computational Mechanics



"Multiscale Computational Mechanics for Material and Structures"
18-19-20 September 2002, Cachan, France

Web site

Registration form

Tentative list of speakers

Rensselaer Polytechnique Institute


- Multiscale methods: theory and computation
- Experimental tests and Identification for multiscale modelling
- Multiscale modelling of damage and fracture
- Verification and validation of multiscale models
- Engineering applications
- Hierarchical multiscale models
- Adaptive parallel computational strategies for multiscale problems
- Multiscale multiphysics problems
- Coupled continuum - atomistic models.


Laboratoire de Mecanique et Technologie
ENS de Cachan/CNRS/Paris 6 University
61 avenue du President Wilson
94235 CACHAN CEDEX France
Phone: (33) 1 47 40 22 41/22 53
Fax: (33) 1 47 40 27 85

Civil Engineering, Mechanical Engineering,
Aerospace Engineering Center
Rensselaer Polytechnique Institute
NY 12180 TROY - USA
Phone: 1 518 276 6191
Fax: 1 518 276 4833


From: Len Freeman (Maths/CNC)" <>
Date: Tue, 07 May 2002 13:55:03 +0100
Subject: MSc in Computational Science at Manchester

University of Manchester
MSc Course in Computational Science

The Department of Computer Science at the University of Manchester has
secured funding from the EPSRC to support a Masters Training Package
in Computational Science. This is a one-year modular MSc course.

The aim of the MSc in Computational Science is to attract high quality
graduates in the Physical Sciences and Engineering who already have an
understanding of the role of (Partial) Differential Equations in
modelling Physical Problems. For most practical problems of interest
Numerical Simulation is the only tool available for solving such
Differential Equations. The objective of this programme is to develop
an appreciation of the range of issues - algorithmic, software
(including visualisation of the results) and hardware - that arise in
the numerical solution of practical Partial Differential
Equations. The programme aims to fulfil a need for graduates who
understand both the Physics and/or Engineering of a practical problem
as well as the Computational Science issues that arise in the
development of high performance numerical solutions.

This MSc course is supported by the EPSRC with (partial) funding
available to support UK or EU students. Applicants should have, or
expect to gain, a good honours degree, or equivalent, in any
Mathematically-based Science or Engineering discipline. There is also
provision for applicants wishing to embark on the course part-time.
Detailed information concerning the contents, prerequisites and the
application procedure for this MSc course can be found at:

Early application is strongly advised. For further information please
Dr Len Freeman, Department of Computer Science,
University of Manchester,
Manchester, M13 9PL
(Tel: +44 161 275 7190, FAX: +44 161 275 6204,


From: Esmond Ng <>
Date: Fri, 10 May 2002 23:33:43 -0700
Subject: Staff Position at Lawrence Berkeley National Laboratory

The Scientific Computing Group in the National Energy Research
Scientific Computing (NERSC) Division ( at the
Lawrence Berkeley National Laboratory (LBNL) has an immediate opening
for a Computer Systems Engineer position. The successful candidate will
engage in research and development in numerical linear algebra that are
adapted to modern high performance computing platforms such as those at
NERSC. Current areas of interest include, but are not limited to,
large-scale eigenvalue calculations and incomplete factorization based
preconditioning techniques for iterative methods. The individual is
expected to interact and collaborate closely with researchers from
application areas that are relevant to DOE missions, particularly those
in the SciDAC Program ( The application areas
include, but are not restricted to, high-energy physics and chemical

For a junior position, the successful candidate will be an active member
of a multi-institutional team whose primary task is to develop and
implement numerical algorithms for large sparse eigenvalue calculations
and/or preconditioning. The individual is expected to interact and
collaborate with researchers from the application areas, as well as
prepare and present progress reports at internal team meetings. For a
senior position, the qualified candidate will perform original research
in large sparse eigenvalue calculations and/or preconditioning, and is
expected to publish results in peer-reviewed journals and present work
at professional meetings. The individual may also be involved in the
preparation and delivery of technical presentations for a variety of
external audiences, such as funding agencies.

To be considered for a junior position, the applicant must have a
thorough knowledge of numerical linear algebra, and must be fluent in a
high-level programming language, such as Fortran 90/95 or C. Experience
with high-performance computing and mathematical software engineering is
preferred. Excellent oral and written communication skills are
desirable. For a senior position, the applicant should demonstrate
experience in carrying out independent research and development. The
individual should be able to facilitate communication with the research
community, as well as with the application areas. A Ph.D. in
mathematics, computer science, or related field or equivalent experience
is desirable.

This is a two-year term appointment with the possibility of renewal.
Classification will depend upon the applicant's level of skills,
knowledge, and abilities. For further details, please visit the LBNL
employment web site ( and search for "014798", or
contact Esmond G. Ng (


From: Joseph Traub <>
Date: Wed, 8 May 2002 11:59:00 -0400 (EDT)
Subject: Contents, Journal of Complexity


Sergei V. Pereverzev and Grzegorz W. Wasilkowski


Linear vs. Standard Information for Scalar Stochastic Differential Equations
Norbert Hofmann, Thomas Muller-Gronbach, and Klaus Ritter

On the L-2 Discrepancy of the Sobol-Hammersley Net in Dimension 3
Gerhard Larcher and Friedrich Pillichshammer

On the Representation of Band Limited Functions Using Finitely Many Bits
Hrushikesh N. Mhaskar

Tractability of Integration in Non-periodic and Periodic Weighted Tensor
Product Hilbert Spaces
Ian H. Sloan and Henryk Wozniakowski

Direct Estimation Of Linear Functionals From Indirect Noisy Observations
Peter Mathe and Sergei V. Pereverzev

Average Case Complexity of Weighted Approximation and Integration over
Leszek Plaskota, Klaus Ritter and Grzegorz W. Wasilkowski

Eugene Allgower, Kurt Georg and Christopher Sikorski


Piecewise Linear Approximation of Smooth Compact Fibers
Eugene L. Allgower and Andrew J. Sommese

Extra-Updates Criterion for the Limited Memory BFGS Algorithm for
Large Scale Nonlinear Optimization
M. Al-Baali

On the Complexity of Exclusion Algorithms for Optimization
Eugene L. Allgower, Melissa Erdmann and Kurt Georg
Verifying Topological Indices for Higher Order-Rank Deficiencies
Jianwei Dian and R. Baker Kearfott

On the Complexity of Isolating Real Roots and Computing with Certainty the
Topological Degree
B. Mourrain, N.M.Vrahatis and J.C. Yakoubsohn

A Two-Dimensional Bisection Envelope Algorithm for Fixed Points
Spencer Shellman and Christopher Sikorski

What Is the Complexity of Volume Calculation?
Arthur G. Werschulz and Henryk Wozniakowski


End of NA Digest