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

- Computing Part of the Spectrum of Large, Sparse Systems
- New book, Mesh Generation
- CAMELot, A Tool for Cellular Automatic
- SIPAMPL, Software for Semi-indefinite Programming Problems
- ParaSails, Parallel Sparse Approximate Inverse Preconditioner
- Scottish Computational Mathematics Symposium
- SIAM/ACM Symposium on Discrete Algorithms
- Computational Methods in Decision-Making & Finance
- Faculty Position at Universitaet Marburg
- Lectureship Position at RMCS, Shrivenham
- Postdoctoral Position at Eindhoven University of Technology
- Research Position at ICASE
- Contents, Lecture Notes in Computational Science and Engineering
- Contents, Optimization Methods and Software

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

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

From: Daniil Sarkissian <sarkiss@math.niu.edu>

Date: Mon, 22 May 2000 17:50:00 -0500 (CDT)

**Subject: Computing Part of the Spectrum of Large, Sparse Systems**

Dear Readers of NA-Digest,

I would very much appreciate if you could give me some references for

the state-of-the art methods to solve the following "partial" eigenvalue

problem for large and sparse matrices. I would like to check them

against the method that I try to develop for my Ph.D. thesis.

Complicated methods are OK, especially if computer codes are available.

Problem:

Find the invariant subspace (or eigenvalue-eigenvector pairs) of the

large and sparse matrix A that correspond to the eigenvalues in the

given region.

For example:

Given "c" and "b > 0";

Find all "a" and "x =!= 0" such that

A*x = a*x (a - eigenvalue, x - eigenvector)

and

| a - c | < b or | a - conjugate(c) | < b

(c and b define the region)

Thank you very much in advance.

Best Regards,

Daniil Sarkissian

Department of Mathematics, Northern Illinois University, DeKalb, IL, 60115

e-mail: sarkiss@math.niu.edu, phone: (815) 754-5787, fax: (815) 753-1112

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

From: Pascal Frey <Pascal.Frey@inria.fr>

Date: Mon, 22 May 2000 10:58:03 +0200

**Subject: New book, Mesh Generation**

A new book on mesh generation and adaptation.

Mesh Generation

Pascal Jean FREY and Paul Louis GEORGE

Hermes Science Publishing, Oxford, Paris

814 pages, 23 chapters

The aim of this book is to provide a comprehensive survey of the

different algorithms and data structures useful for triangulation

and meshing construction. In addition, several aspects will also be

described, for instance mesh modification tools, mesh evaluation

criteria, mesh optimization, including even adaptive mesh

construction as well as parallel meshing techniques.

book will be available in stores or can be ordered directly from

the editor:

http://www.editions-hermes.fr

Pascal

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

From: Carlo Nardone <cmn@roma.quadrics.com>

Date: Wed, 24 May 2000 21:31:43 +0200

**Subject: CAMELot, A Tool for Cellular Automatic**

In the framework of the ESPRIT-IV programme the European Commission

has sponsored COLOMBO, a R&D project which resulted in the development

of a tool, CAMELot, for the development, parallelization and monitoring

of Cellular Automata and similar models. More info about the project

are available at the URL http://www.epcc.ed.ac.uk/~colombo or

http://www.epcc.ed.ac.uk/industry/COLOMBO/index.html

The European Commission requested the project partners to assess the

market potential for the CAMELot tool. Anyone willing to collaborate

can point its browser to the URL

http://www.epcc.ed.ac.uk/~colombo/questionnaire.html

to fill-in the on-line questionnaire form.

The project partners thank in advance for the collaboration.

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

From: Antonio Ismael F. Vaz <aivaz@aleixo.ci.uminho.pt>

Date: Thu, 25 May 2000 15:07:40 +0100

**Subject: SIPAMPL, Software for Semi-indefinite Programming Problems**

We would like to announce the SIPAMPL software package. The SIPAMPL package

is an extension to the AMPL modelling language environment that allows the

codification of semi-infinite programming (SIP) problems.

SIPAMPL includes:

1) A database with 71 SIP problems, coded in AMPL;

2) An interface that allows the connection between AMPL and any SIP solver

(namely, for the MATLAB solver);

The SIPAMPL software package is freely available as well as a Technical Report

that explain on how one should use the database and the interface.

See http://www.eng.uminho.pt/~dps/aivaz .

Feel free to forward this message to anyone potentially interested in this

subject.

Comments and contributions are welcome.

Sincerely,

Ismael Vaz

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

From: Edmond Chow <chow@iona.llnl.gov>

Date: Fri, 26 May 2000 15:07:51 -0700 (PDT)

**Subject: ParaSails, Parallel Sparse Approximate Inverse Preconditioner**

ParaSails, Parallel Sparse Approximate Inverse Preconditioner

available at http://www.llnl.gov/casc/parasails

ParaSails is a parallel sparse approximate inverse preconditioner for

the iterative solution of large, sparse systems of linear equations.

ParaSails has been used to solve finite element elasticity problems

inside a Lawrence Livermore simulation code with more than 4 million

equations on 1000 processors of ASCI Blue-Pacific (IBM SP). It has

also been demonstrated on anisotropic diffusion problems with 216

million equations.

ParaSails uses least-squares (Frobenius norm) minimization to compute a

sparse approximate inverse. The sparsity pattern of the approximate

inverse is the pattern of a power of a sparsified matrix. ParaSails

also uses a post-filtering technique to reduce the cost of applying the

preconditioner. The pattern of the preconditioner can be reused to

generate preconditioners for different matrices in a sequence of linear

solves.

ParaSails solves symmetric positive definite problems using a

factorized preconditioner. ParaSails can also solve general

(nonsymmetric and/or indefinite) problems with a nonfactorized

preconditioner. The software available to be downloaded includes

parallel CG and GMRES solvers, a parallel matrix data structure and a

test driver. It is implemented with MPI message passing.

The software is freely available for research and development

purposes. Commercial use is possible with a licensing agreement.

The software avilable at: http://www.llnl.gov/casc/parasails

Edmond Chow

Center for Applied Scientific Computing, LLNL

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

From: Dave Sloan <dms@maths.strath.ac.uk>

Date: 24 May 2000 12:27:41 +0100

**Subject: Scottish Computational Mathematics Symposium**

SCOTTISH COMPUTATIONAL MATHEMATICS SYMPOSIUM 2000

First Announcement

Friday 22 September 2000 at University of Strathclyde

The ninth annual Scottish Computational Mathematics Symposium will

be held at the University of Strathclyde, Glasgow, on Friday 22

September 2000.

SPEAKERS

Roger Fletcher University of Dundee

Ben Leimkuhler University of Leicester

Gabriel Lord Heriot-Watt University

Sean McKee University of Strathclyde

Endre Suli University of Oxford

Details will be circulated later.

REGISTRATION FEE: 20 pounds (covers tea/coffee/lunch)

CONTACTS:

Dugald Duncan Heriot-Watt University dugald@ma.hw.ac.uk

Dave Sloan University of Strathclyde d.sloan@strath.ac.uk

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

From: Darrell Ross <ross@siam.org>

Date: Thu, 25 May 2000 13:06:28 -0400

**Subject: SIAM/ACM Symposium on Discrete Algorithms**

Twelfth Annual SIAM/ACM Symposium on Discrete Algorithms (SODA)

Important Dates:

Long form abstracts in electronic form are due July 10, no later than 5:00

PM EDT. Instructions and information about electronic submissions will be

posted later at: http:/www.siam.org/meetings/da01/ .

Short form abstracts in electronic form are due no later than August 2,

5:00 PM EDT. Instructions and information about electronic

submissions will be posted later at: http:/www.siam.org/meetings/da01/ .

To know more about this conference, visit:

http://www.siam.org/meetings/da01/

or contact SIAM by:

e-mail: meetings@siam.org

phone: 215-382-9800

fax: 215-386-7999

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

From: Paolo Foschi <paolo.foschi@info.unine.ch>

Date: Fri, 26 May 2000 18:35:02 +0200

**Subject: Computational Methods in Decision-Making & Finance**

Computational Methods in Decision-Making & Finance (CMDMF)

16th of August 2000, Neuchatel, Switzerland

http://iiun.unine.ch/matrix/seminars/CMDF2000/3ec2000b.html

This meeting will emphasize both practical and theoretical,

state-of-the-art applications of computational technology to solve

financial and decision-making problems. The panel of speakers is made

up of internationally respected experts:

o Anna Nagurney, University of Massachusetts, USA

Financial networks

o Berc Rustem, Imperial College, London, UK

Worst-case design algorithms and risk management

o Stavros Siokos, Portfolio Construction Group, Salomon Smith Barney,

London, UK

Portfolio construction, index tracking and optimization:

computational issues

o Patrick Burns, Global Quantitative Research, Salomon Smith Barney,

London, UK

Multivariate GARCH: solving hard maximum likelihood problems using a

genetic algorithm

o Manfred Gilli and Evis Kellezi, Department of Econometrics,

University of Geneva, Switzerland

A global optimization heuristic for portfolio choice with VaR and

expected shortfall

o Olivier V. Pictet, Dynamic Asset Management, Geneva, Switzerland;

Oliver Masutti and Gilles Zumbach, Olsen & Associates, Zurich,

Switzerland

Volatility forecasting using genetic programming

o Cyril Godart, Paribas, London, UK

Swarm and huscarls: a case-study in high-performance computing in

financial institutions

o Panos M. Pardalos University of Florida, USA; and

M.G.C. Resende AT&T Labs Research, USA.

Parallel Metaheuristics for Combinatorial Optimization Problems

(pending)

Authors wishing to present a paper are invited to submit an abstract

(maximum two pages), by e-mail to:

Erricos John Kontoghiorghes

Institut d'informatique,

Universite de Neuchatel,

Emile-Argand 11,

CH-2007 Neuchatel,

Switzerland.

E-mail: erricos.kontoghiorghes@info.unine.ch

Fax: +41 32 718 27 01

Tel: +41 32 718 27 38

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

From: Bernhard Schmitt <schmitt@Mathematik.Uni-Marburg.de>

Date: Mon, 22 May 2000 17:21:57 +0200 (MET DST)

**Subject: Faculty Position at Universitaet Marburg**

Philipps-Universitaet Marburg, Germany

Am Fachbereich Mathematik und Informatik ist zum 01.10.2001 eine

Professur (C4) fuer Mathematik (Numerik)

zu besetzen. Erwuenscht ist, dass die Stelleninhaberin/der Stelleninhaber

ein aktuelles Gebiet der Numerischen Mathematik (z.B. Numerik von

Differentialgleichungen, nicht-lineare Probleme, numerische

Approximation) in Forschung und Lehre vertritt.

Erwartet wird, dass sich die Stelleninhaberin/ der Stelleninhaber an

der Grundausbildung im Fachbereich sowie der Ausbildung im Hauptstudium

Mathematik und Wirtschaftsmathematik engagiert beteiligt.

Fuer die Ernennung zum Professor oder zur Professorin sind

paedagogische Eignung und - als Nachweis der Befaehigung zu

wissenschaftlicher Arbeit - neben der Promotion zusaetzliche

wissenschaftliche Leistungen (z.B. Habilitation) erforderlich. Die

paedagogische Eignung ist durch geeignete Unterlagen, beispielsweise

ueber durchgefuehrte Lehrveranstaltungen, Leistungen in der Lehre,

hochschuldidaktische Aktivitaeten etc. nachzuweisen.

Der Frauenfoerderplan der Philipps-Universitaet verpflichtet zur

Erhoehung des Frauenanteils. Frauen sind deshalb ausdruecklich zur

Bewerbung aufgefordert. Unter den Voraussetzungen nach Paragraph 85a HBG

ist grundsaetzlich eine Reduzierung der Arbeitszeit moeglich.

Schwerbehinderte Bewerber/Bewerberinnen werden bei gleicher Eignung

bevorzugt.

Bewerbungen sind mit den ueblichen Unterlagen bis zum 07. Juli 2000 an

den Praesidenten der Philipps-Universitaet, Biegenstrasse 10,

D-35032 Marburg, zu richten.

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

From: Joyce Aitchison <J.M.Aitchison@rmcs.cranfield.ac.uk>

Date: Thu, 25 May 2000 20:17:33 +0100

**Subject: Lectureship Position at RMCS, Shrivenham**

Applied Mathematics & Operational Research Group

Cranfield University - Royal Military College of Science, Shrivenham.

LECTURER in COMPUTATIONAL MATHEMATICS or STATISTICS

Applications are invited for a post of Lecturer in Computational Mathematics

or Statistics in the Applied Mathematics and Operational Research Group of

Cranfield University at the Royal Military College of Science, Shrivenham.

The group is responsible for all the Mathematics, Statistics & Operational

Research teaching within RMCS which includes input to all the undergraduate

and postgraduate degrees and various short courses. The group also runs a

suite of modular full and part-time MSc degrees.

The appointee will contribute to the teaching of mathematics or statistics

to the specialist MSc and short course programmes. He/she will also provide

service teaching to non-specialist courses and must be prepared to work with

a wide range of students.

Applicants should have an established record in an appropriate research area

or clear research potential. The successful candidate will be expected to

develop his/her research activities and to supervise postgraduate students.

The group has an active research programme including EPSRC and other

externally and internally funded PhD students. The current research

activities of the group include industrial mathematics, numerical methods,

scientific computing and applied statistics.

Applications are invited from candidates with compatible research interests,

although experience relevant to the computational mathematics or statistics

activities of the group would be particularly welcome. Flexibility and

willingness to work with other members of staff and a variety of students

are essential.

Informal enquiries may be made to Joyce Aitchison on (01793) 785276 or email

J.M.Aitchison@rmcs.cranfield.ac.uk or see our website at

http://www.rmcs.cranfield.ac.uk/departments/dois/amorg/

Application forms and further details may be obtained from the Personnel

Office, Cranfield University, RMCS Shrivenham, Swindon, SN6 8LA. Tel:

(01793) 785758 quoting reference: R/DIS 21.

Completed applications should be sent to the Personnel Office before the

closing date of 23 June 2000.

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

From: J. Molenaar <J.Molenaar1@tue.nl>

Date: Tue, 23 May 2000 15:57:59 +0200

**Subject: Postdoctoral Position at Eindhoven University of Technology**

At the department of Mathematics and Computing of the Eindhoven University

of Technology is a vacancy for a

POST-DOC position Applied Mathematics

(V 32848)

THE PROJECT

Eindhoven University of Technology participates in the European Union

project "Postpone Polymer Processing Instabilities (3PI)". The 3PI is

a common enterprise of 10 partners: 5 universities and 5 companies, all

active in polymer research.

When polymer melts are extruded, the extrudate may show instable

behaviour:

instead of a smooth appearance surface distortions or even volumetric

distortions of the extrudate are observed if the extrusion rate is

increased.

These instabilities are commonly referred to as 'polymer melt fracture'.

They phenomena seriously limit production rates. The object of the 3PI

project is to develop models which link to the appearance of the flow

instabilities, the viscoelastic properties of the polymer, the boundary

conditions at the die wall and the processing conditions.

The contribution of Eindhoven University is to provide modelling tools

from

mathematics. The Eindhoven research focusses on the modelling of the

interaction between the melt and the die wall.

It is believed that the complexity of this interaction is one of the main

reasons that melt fracture is not yet fully understood since its first

detection about half a century ago.

Eindhoven is located in the south of The Netherlands. Polymers are a topic

of intense research at this university. The present project requires

regular

contacts with the other 3 PI partners. Funding for travelling and visiting

conferences is available.

Information about Eindhoven University and Department of Mathematics and

Computing Science can be found on the web site:

http://www.win.tue.nl/~anwww

THE CANDIDATE

The candidate should have a Ph.D. degree in mathematics or physics or

have a similar level, and be able and motivated to develop and apply

mathematical models in the field of polymer melt dynamics (rheology).

Experience in fluid mechanics and/or reptation theories is an advantage.

Numerical evaluation of the developed models is part of the task.

EMPLOYMENT

The candidate gets a temporary position as a staff member for at most 3.5

years. The salary is according to the 'collective employment agreement

dutch universities'. Depending upon education and experience the salary

will be up to a maximum of Dfl. 7694,00 gross a month.

INFORMATION

Concerning the project: Prof. Dr. J.Molenaar

Tel. : 00-31-40-2474757(2474760)

E-mail: j.molenaar1@tue.nl

Internet: http://www.win.tue.nl/~jaapm/vacancies

About the job conditions: W.C.J. Verhoef, Personnel Officer,

tel. 00-31-40-2472321. E-mail: w.c.j.verhoef@tue.nl

HOW TO APPLY

You can send a written application, together with your CV and a list of

publications, before June 22th, 2000 to drs S. Udo, managing director of

the faculty of Mathematics and Computing Science, Eindhoven University of

Technology, HG 6.22, P.O.Box 513, 5600 MB Eindhoven, mentioning the number

of the vacancy V32848.

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

From: Manuel Salas <salas@icase.edu>

Date: Tue, 23 May 2000 14:31:09 -0400

**Subject: Research Position at ICASE**

Senior Research Position in Computer Science

ICASE

Hampton, VA

The Institute for Computer Applications in Science and Engineering

(ICASE) is seeking a senior researcher to lead the institute's

Computer Science research effort. In recent years, CS research at

ICASE has focused on enhancing the performance and utility of parallel

and distributed computing systems. This research is driven by the

needs of multidisciplinary scientific applications of interest to NASA

and related industries. In the area of parallel computing, the research

focuses on the development of algorithms, methodologies and tools for

efficient porting of scientific applications to contemporary and future

parallel architectures with deep and heterogeneous memory hierarchies.

In the area of distributed computing, the aim of the research is to

utilize new technologies such as the Web and Java, to provide integrated

environments for scientists and engineers to exploit a distributed

network of heterogeneous resources for large-scale distributed

scientific applications.

ICASE is seeking a person who can provide the technical and managerial

expertise to lead and enhance the program. This position will require

conducting in-house research on problems of interest to NASA in addition

to forming leading edge collaborative research efforts with NASA

scientists and engineers and researchers from universities and national

labs.

Compensation will be commensurate with experience and qualifications.

U.S. citizens or permanent resident aliens are preferred, but foreign

nationals are not excluded. Applications from women and minorities are

also encouraged.

ICASE is a non-profit research institute located at the NASA Langley

Research Center in Hampton, Virginia. The institute conducts research

in a wide range of topics of interest to NASA. In addition to computer

science, research topics include formal methods, fluid mechanics,

applied and numerical mathematics, structures and materials and

multidisciplinary optimization. ICASE researchers have access to an

in-house PC-cluster comprised of 64 compute nodes and large data storage.

They also have access to all of NASA's parallel computers and many other

parallel systems at government labs.

Additional information about ICASE and its research programs is

available via the World Wide Web at "http://www.icase.edu".

Please send resumes to:

Director

ICASE, MS 132C

NASA Langley Research Center

Hampton VA 23681-2199

U.S.A.

or by e-mail to:

positions@icase.edu

ICASE is an equal opportunity employer.

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

From: Thanh-Ha LeThi <LeThi@Springer.de>

Date: Thu, 25 May 2000 14:42:52 +0200

**Subject: Contents, Lecture Notes in Computational Science and Engineering**

Lecture Notes in Computational Science and Engineering, Vol. 12: U. van

Rienen "Numerical Methods in Computational Electrodynamics - Linear Systems

in Practical Applications"

2000. DM 149,-; Softcover ISBN 3-540-67629-5

This interdisciplinary book deals with the solution of large linear systems

as they typically arise in computational electrodynamics. It presents a

collection of topics which are important for the solution of real life

electromagnetic problems with numerical methods - covering all aspects

ranging from numerical mathematics up to measurement techniques. Special

highlights include a first detailed treatment of the Finite Integration

Technique (FIT) in a book - in theory and applications, a documentation of

most recent algorithms in use in the field of Krylov subspace methods in a

unified style, a discussion on the interplay between simulation and

measurement with many practical examples.

Lecture Notes in Computational Science and Engineering, Vol. 15: A. Frommer,

T. Lippert, B. Medeke, K. Schilling (Eds.) "Numerical Challenges in Lattice

Quantum Chromodynamics - Joint Interdisciplinary Workshop of John von

Neumann Institute for Computing, J=FClich and Institute of Applied Computer

Science, Wuppertal University, August 1999"

2000. DM 129,-; Softcover ISBN 3-540-67732-1

The book with contributions from the joint interdisciplinary workshop covers

important numerical bottleneck problems from lattice quantum chromodynamics:

1) The computation of Green's functions from huge sparse linear systems and

the determination of flavor-singlet observables by stochastic estimates of

matrix traces can both profit from novel preconditioning techniques and

algebraic multi-level algorithms. 2) The exciting overlap fermion

formulation requires the solution of linear systems including a matrix sign

function, an extremely demanding numerical task that is tackled by

Lanczos/projection methods. 3) Realistic simulations of QCD must include

three light dynamical quark flavors with non-degenerate masses. Algorithms

using polynomial approximations of the matrix determinant can deal with this

situation. The volume aims at stimulating synergism and creating new links

between lattice quantum and numerical analysis.

Springer Series in Computational Mathematics Vol. 28: Zhen Mei "Numerical

Bifuraction Analysis for Reaction-Diffusion Equations"

2000. DM 159,-; Hardcover ISBN 3-540-67296-6

This book provides the readers numerical tools for a systematic analysis of

bifurcation problems in reaction-diffusion equations. Emphasis is put on

combination of numerical analysis with bifurcation theory and application

to reaction-diffusion equations. Many examples and figures are used to

illustrate analysis of bifurcation scenario and implementation of numerical

schemes. The reader will have a thorough understanding of numerical

bifurcation analysis and the necessary tools for investigating nonlinear

phenomena in reaction-diffusion equations

For further information, please contact

Dr. Martin Peters

Senior Mathematics Editor Phone: *49-6221-487 185

Springer-Verlag FAX: *49-6221-487 355

Tiergartenstrasse 17 mailto:Peters@Springer.de

D-69121 Heidelberg, Germany http://www.springer.de/math/peters.html

Visit our Web site at http://www.springer.de

or http://www.springer-ny.com

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

From: Oleg Burdakov <olbur@mai.liu.se>

Date: Fri, 26 May 2000 20:40:59 +0200 (MET DST)

**Subject: Contents, Optimization Methods and Software**

Table of Contents

Optimization Methods and Software (OMS)

Volume 13, Number 3 (June, 2000)

M. Al-Baali

Extra updates for the BFGS method

159-179

Dong-Hui Li and Masao Fukushima

A derivative-free line search and global convergence of Broyden-like

method for nonlinear equations

181-201

M. Mongeau, H. Karsenty, V. Rouze, J.-B. Hiriart-Urruty

Comparison of public-domain software for black-box global optimization

203-226

Forthcoming papers and complete table of contents for the journal OMS:

http://www.mai.liu.se/~olbur/OMS.contents

Instructions for authors, subscription information, free sample copies:

http://www.gbhap.com/Optimization_Methods_Software/

Latex style files:

http://www.mai.liu.se/~olbur/STYLES/

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

End of NA Digest

**************************

-------