NA Digest Saturday, March 22, 1997 Volume 97 : Issue 12

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: Nicolas Robidoux <>
Date: Fri, 21 Mar 1997 21:21:39 GMT
Subject: Question about Intellectual Property Rights

A friend of mine is thinking of taking a tenure track position at a US
department of Mathematics or Computer Science after having worked at a
European institute for a half decade.

He has developed a software package which, although not commercializable
in its current form, may become so in a few years. A similar, but more
restricted, package was licensed to a major European Engineering modeling

Is it true that some universities/colleges automatically have partial
ownership of commercial software or products developed by faculty and
graduate students while at their employ? If so, what is usually
considered benign, and what is "pushing it?" Any advice on what and
who to ask before signing one's life away?

Replies sent directly to me will be summarized. Indicate if you want
to remain anonymous.

With thanks,

Nicolas Robidoux
Albuquerque Resource Center of the
High Performance Computing and Research Center
Albuquerque, NM


From: Maurizio Motolese <motolese@leland.Stanford.EDU>
Date: Mon, 17 Mar 1997 16:46:50 -0800 (PST)
Subject: Software to Solve Large Systems of Non-linear Equations

We are graduate students at the departement of Economics at Stanford University.
Our research in general equilibrium economics requires us to solve systems of
upwards of 500 nonlinear equations with the property that the jacobian will be
singular in some subspaces not near the root. We have used routines employing
tensor product terms in the approximation for a Newton method to solve a
smaller system. We would greatly appreciate any suggestions you might have
concerning an appropriate software routine for solving such large systems.
(FORTRAN preferred but not necessarily).

Thank you very much in advance for your help!

Maurizio Motolese

Stan Black

Dept. of Economics
Stanford University
Stanford, CA 94305-6072


From: Ken Turkowski <>
Date: Tue, 18 Mar 97 00:37:58 -0800
Subject: Re: Tesselation of Sphere

In last week's NA Digest, Siang Peng Oh <peng@astro.Princeton.EDU> asked:

> Does anyone know of the optimal way of tiling the sphere into a large
> number (N~10**6) of approximately equal area, equal angular size tiles?
> I am working on the Cosmic Microwave Background, and we would
> like to consider the optimal way of pixelizing the sky. Currently we
> inscribe one of the Platonic regular solids onto the sphere and pixelize
> the solid, but for various reasons it would be nice to have a grid which
> is as regular as possible in (theta,phi). I know an exact solution is not
> possible for N>20, but wonder if there is a close to optimal solution for
> integrating over the sphere.

Try this easy one:
Separate the sphere into octants. Subdivide each equilateral triangle
into fourths by bisecting the midpoints of each side:

. . . .
. . . .
. . . .
. . . .
. .
. .
. .
. .

The octant subdivision yields 2^3, and each subsequent subdivision yields
2^2. Subdividing the octants 8 times will give you 2^19 or 0.5E+6

Ken Turkowski Apple Computer, Inc. Cupertino, CA 95014
email: vox: (408) 974-6699 fax: (408) 974-8414


From: Jack Dongarra <>
Date: Thu, 20 Mar 1997 09:09:22 -0500
Subject: The NetSolve Project

The NetSolve Project

The first public release of NetSolve is now available. NetSolve is a
network enabled solver that allows users to access computational
resources, such as hardware and software, distributed across the network.
The development of NetSolve was motivated by the need for an easy-to-use,
efficient mechanism for accessing computational resources remotely.
Ease of use is obtained as a result of different interfaces, such as
Fortran, C, and Matlab; good performance is ensured by a load-balancing
policy that enables NetSolve to allocate computational resources as
efficiently as possible. NetSolve offers the ability to look for
computational resources on a network, choose the best one available,
solve a problem (with retry for fault-tolerance), and return the answer
to the user.

The software and additional information is available from our home page:

This page provides information about the software release, plus papers and
manuals on NetSolve. If you have any questions, please mail us at

Brief Description of the System
NetSolve has three components: the client, which can be either a
user program or a user interacting with one of the NetSolve interfaces;
the NetSolve agent; and the pool of NetSolve resources. The entry
point into the NetSolve system is the client sending a problem request
to the agent. The agent analyzes this request and chooses a computational
resource. The problem and its input data are then sent to the chosen
NetSolve resource. The problem is solved by the appropriate software
package, and the result is sent back to the client. The system can be
set up on an intranet or over the internet. We have set up an agent and
a number of computational servers that can be used here in Tennessee.

Contacts and Support
NetSolve is located at This location
contains the source code distributions. Questions and comments can be
directed via e-mail to

Henri Casanova and Jack Dongarra


From: John Mathews <>
Date: Tue, 18 Mar 1997 07:46:49 -0800 (PST)
Subject: Conformal Mapping Software

Examples of Conformal Mappings illustrated with the Computer Software F(Z)
are now available at the Mathematics Archives WWW site:

The collection titled "COMPLEX ANALYSIS: F(Z) Files" is located at:

To obtain it just download the file CA.ZIP [103 KB]. It contains of the
complete collection of examples for the new textbook: COMPLEX ANALYSIS: for
Mathematics & Engineering 3rd Ed., 1997, by John Mathews and Russell Howell
Jones and Bartlett Publishers, Inc. ISBN: 0-7637-0270-6 More information
for textbook is located at:

COMPLEX ANALYSIS: Computer Software Supplements for MAPLE and Mathematica
are being developed at this time and one for Matlab is being planned. If you
have an interest in them you should contact me directly.


John H. Mathews
Department of Mathematics
California State Univ. Fullerton
Fullerton, CA 92834 USA



From: Alan Karp <>
Date: Mon, 17 Mar 1997 11:38:17 -0800
Subject: Gordon Bell Prize for 1997

This is the 10th year of the Gordon Bell Prize for parallel
processing. Although he originally stated that he would fund the
prize for 10 years, Gordon has agreed to continue to offer the prize
until someone sustains at least 1 Tf/s on a real application. The
rules for 1997 are attached. Note that criteria other than raw
performance are considered.

The 1997 Gordon Bell Prizes

The Gordon Bell Prizes recognize achievements in large-scale
scientific computing. Entries for the next Prize are due on 1 May
1997, and finalists will be announced by 30 June 1997. Pending
approval by the Supercomputing '97 program committee, finalists will
be invited to present their work at a special session of that meeting
in November 1997. Winners and honorable mentions will be announced
following the presentations.

The 1997 prizes will be given for work in the following categories:

1. Performance: The entrant will be expected to convince the judges
that the submitted program is running faster than any other
comparable engineering or scientific application. Suitable evidence
will be the megaflop rate based on actual operation counts or the
solution of the same problem with a properly tuned code on a machine
of known performance, such as a Cray Y-MP. If neither of these
measurements can be made, the submitter should document the
performance claims as well as possible.

2. Price/performance: The entrant must show that the performance of
the application divided by the list price of the smallest system
needed to achieve the reported performance is better than that of
any other entry. Performance measurements will be evaluated as for
the performance prize. Only the cost of the CPUs, memory, and any
peripherals critical to the application need be included in the
price. For example, if the job can be run on diskless compute
servers, the cost of disks, keyboards, and displays need not be

3. Compiler parallelization: The combination of compiler and
application that generates the most speed-up will be the winner.
Speed-up will be measured by dividing the wall clock time of the
parallel run by that of a good serial implementation of the same
job. These may be the same program if the entrant can convince the
judges that the serial code is a good choice for a uniprocessor.
Compiler directives and new languages are permitted. However, anyone
submitting an entry in other than a standard, sequential language
will have to convince the judges that the parallelism was detected
by the compiler, not by the programmer.

There are some general conditions:

1. The submitted program must have utility; it must solve a problem
that is considered a routine production run, such as making daily
weather predictions or solving an important engineering or
scientific problem. It should not be a contrived or experimental
problem that is intended just to show high speed-up.

2. Entrants in the price/performance category must demonstrate that
the machine they used has real utility. (No fair picking up a few
used Z-80s for $1 each.) Only list prices of components should be
used. If the machine is not on the market, the entry is probably not
eligible although the judges will consider any reasonable estimate
of the price.

3. One criterion the judges will use for all categories is how much
the entry advances the state of the art of some field. For example,
an entry that runs at 100 Gflops but solves a problem in a day that
previously took a year might win over an entry that runs at 150
Gflops solving a more mundane problem. Entrants who believe their
submission meets this criterion are advised to document their claims

4. In all cases the burden of proof is on the contestants. The
judges will make an honest effort to compare the results of
different programs solving different problems running on different
machines, but they will depend primarily on the submitted material.

Contestants should send a three or four page executive summary to
Marilyn Potes, IEEE Computer Society, 10662 Los Vaqueros Circle, Los
Alamitos, CA 90720-2578 before 1 May 1997.

Alan Karp
Hewlett-Packard Labs
1501 Page Mill Road
Palo Alto, CA 94304
(415) 857-6766, (415) 813-3381 (fax)


From: I.Egorov <>
Date: Thu, 20 Mar 1997 13:32:53 +0300
Subject: New Optimization Technology

We invite to cooperation about the new Optimization Technology.
Our technology is designed for multimeasure (up to 100 and more
variables) optimization practical problems with continuous, breaking
nondifferentiative and stochastic goal functions.
I send you a short list of our optimization technology.

At the moment the raise of effectiveness of the complex technical
systems (such as gas-turbine engines, flying vehicles etc.) is reached
by introduction of the new manufacturing technologies, by application of
the new conceptual solutions etc. However, for majority of engineering
systems it is possible to achieve an essential raise of their
effectiveness at the expense of multiparameter (50 and more variables)
and multicriteria optimization and reduction of cost of modernization
For this purpose the unique technology for optimization investigations
has been developed. The technology is oriented to researches, analysis,
search of ways to increase the efficiency and development forecasting of
complex technical systems by means of
- optimum designing of the separate elements and system as a whole;
- optimum matching of the elements, included in the systems;
- determination of the optimum control laws.
during their designing, operational development and modernization. The
technology is based on a new method of indirect optimization on the
basis of self-organizing - IOSO, which has been developed by DSc
The distinctive features of the technology are:
- multicriteria optimization possibility for the problems of large
dimensionality (up to 200 and more variables), that allows to reach the
object's efficiency increasing, which is up to 2...7 times higher than
under traditional optimization methods use.
- low costs for search of the optimum decision (reduction of a time
required for search of the decision up to 20 ...150 times depending on
complexity and dimensionality of a problem;
- multicriteria optimization for the problems in stochastic statement
(up to 100 variables), with a complex topology of the goal functions and
with a plenty of the constraints (the analogues of the decision of
similar problems are unknown);
- decision opportunity for the optimization problems of various classes,
including stochastic, multiextreme and nondifferentiated.
A distinctive feature of the given technology is its high effectiveness
in search for an optimum solution when investigating into technical
systems, modeled at high levels of complexity and hierarchy including
the last achievements in mathematical modeling (2-D and 3-D problems).
In an engine-building the given technology was used for the solution of
the following practical problems (for the "Aviadvigatel",Inc; "Lyulka
Saturn",Inc; SSTC "NK-Engines"; SNECMA, "AutoVAZ"):
1. Determination of the perfection level and analysis of the
possibilities to improve the characteristics of subsonic and supersonic
variable and non-variable axial flow compressors on the basis of the 2-D
axisymmetrical mathematical model (number of variables - up to 99;
number of constraints - up to 220; number of optimization criteria - up
to 4).
2. Optimum control of aircraft gas-turbine engines for unsteady
operational modes (number of variables - up to 216; number of
constraints - up to 26).
3. Matching of flying vehicle and gas-turbine engine, and definition its
optimal control laws (number of variables - up to 70; number of
constraints - up to 8; number of optimization criteria - up to 2).
4. Reduction of the negative influence of the compressor flow path
erosion on the engine's characteristics.
5. Optimum calibration of the microprocessor control systems of a
automobile engine directly on a test stand for ensuring of a minimum of
the fuel consumption at a given emission level.
6. Definition of the parameters and structure of a covers of the engine
units for ensuring the given optical, thermal and strength
characteristics (number of variables - up to 20, number of constraints -
up to 50).
The technology is invariant as to the objects to be investigated and can
be easily adapted to using mathematical models of different complexity
level applied by design companies when solving a large series of
practical problems in various fields of science and engineering
(mechanical engineering, medicine, chemistry etc).
The more detailed technology description and the examples of its usage
for the aviation engines are cited in the following papers:
1. Beknev V.S.,Egorov I.N., Talyzina V.S. "Multicriteria Design
Optimization of the Multistage Axial Flow Compressor." 5-th ASME,
"COGEN-TURBO-V" Budapest, Hungary, 1991.
2. Egorov,I.N. "Optimization of a Multistage Axial Compressor.
Stochastic Approach", ASME , 92-GT-163, 1992.
3. Egorov I.N.,Kretinin G.V., "Multicriterion Stochastic Optimization of
Axial Compressor." ASME, COGEN-TURBO VI, Houston, USA, 1992.
4. Egorov I.N. "Deterministic and Stochastic Optimization of Variable
Axial Compressor." ASME, 93-GT-397. 1993.
In case of your interest we are ready to present you the more complete
information and to decide your optimization problem as the demonstration
of our possibilities.

Studentcheskaya 42, off.109, Moscow, 121165, Russia
tel/fax (7 095) 249-1963
DSc. I.Egorov


From: Hans-Christian Hege <>
Date: Thu, 20 Mar 1997 11:39:45 +0100
Subject: VideoMath Festival, Call for Videos

VideoMath Festival
at the
International Congress of Mathematicians 1998 (ICM'98)
Berlin (Germany)
August 18-27, 1998

The International Congresses of Mathematicians, taking place about
every four years since 1897, belong to the most important mathematical
events in the world. One distinguishing feature, among others, is the
award of the Fields Medals and the Nevanlinna Prize (the "mathematical
Nobel Prizes") during the opening ceremony.

The ICM'98 will take place in Berlin, Germany, from August 18 to 27,
1998. It will be accompanied by a number of cultural events. One of
these events will be the VideoMath Festival, a public presentation of
a collection of outstanding mathematical videos.

The festival is planned to attract a broad audience: ICM attendees,
students, teachers, and the proverbial man-in-the-street with an
inclination towards mathematics. Performances are open to the public
and will take place during the period of the conference. The Urania
building, a center for popular science located in downtown Berlin,
will host the VideoMath Festival as well as other cultural events in
the wake of ICM'98.

The collection of videos to be presented will be selected by a program
committee. The video pieces chosen will be integrated into a feature
film of up to two hours length. For more information about the
VideoMath Festival see:


The program committee currently consists of

Thomas Banchoff Mathematics Dept., Brown University,
Providence, USA
Peter Deuflhard Konrad-Zuse-Zentrum fuer Informationstechnik,
Berlin, Germany
George Francis Mathematics Dept. and NCSA,
Urbana-Champaign, USA
Herbert W. Franke Universitaet Muenchen, Germany

David Hoffman Mathematical Sciences Research Institute,
Berkeley, USA
Heinz-Otto Peitgen Fachbereich Mathematik, Universitaet Bremen,
Ulrich Pinkall Fachbereich Mathematik, Technische Universitaet
Berlin, Germany.


The organizing committee for the VideoMath Festival consists of

Hans-Christian Hege Konrad-Zuse-Zentrum fuer Informationstechnik
Berlin, Germany

Konrad Polthier Fachbereich Mathematik, Technische
Universitaet Berlin, Germany


Entries for the VideoMath Festival are encouraged from all areas of
mathematical visualization. The submissions should appeal to a general
but educated public. They should also meet highest standards with
respect to mathematical content, visualization techniques, artistic
design and technical quality. We strongly recommend to make the work
interesting for a broad audience by selecting an appropriate subject,
and by using visual elements, suitable text or narration.

The program committee, a group of internationally renowned
mathematicians and computer graphics experts, will evaluate the entries
and select a number of contributions according to quality and thematic
balance. These will be spliced together for the final tape. To
facilitate adequate judgement, a submission may include additional
notes to the jury on the mathematical content of the work, new
techniques that have been used, or a brief statement describing the
artistic concept of the work.

For more information about submission of videos, see the Web site
mentioned above.


Submission of Video by April 3, 1998
Submission of Final Version by June 5, 1998


From: Bette Byrne <>
Date: Thu, 20 Mar 97 14:59:05
Subject: Meeting Marking the Retirement of Bill Morton


17-18 APRIL 1997


This is a special meeting to honour Bill Morton in the year of his retirement
from the chair of Numerical Analysis at Oxford.

Speakers will include:
Prof C S Morawetz (Courant Institute)
Prof K W Morton (University of Oxford)
Prof B Parlett (University of California)
Prof P Roe (University of Michigan
Dr W L Wood (University of Reading)
Dr J W Barrett (Imperial College)
Dr A Craig (University of Durham)
Dr E S Cli (University of Oxford)
Dr M Rudgyard (University of Oxford)
Dr M Paisley (Staffordshire University)
Dr J A Mackenzie (Strathclyde University)
Dr A Priestley (GeoQuest)
Dr P I Crumpton (University of Oxford)
Dr A J Wathen (University of Oxford)
Dr G Moore (Imperial College)
Dr P Stow (Rolls-Royce)
Dr M J P Cullen (Meteorological Office)
Dr N Nichols (University of Reading)
Dr A K Parrott (University of Greenwich)
Dr P K Sweby (University of Reading
Prof M J Baines (University of Reading)

The meeting will take place in the Computing Laboratory and will begin
mid-morning on Thursday, 17th and finish during the late afternoon on
Friday, 18th. Accommodation for Thursday night will be available at St
John's College. There will also be a Special Dinner in St John's
College on Thursday evening.

If you require any further information please contact:
Bette Byrne
Oxford University Computing Laboratory
Wolfson Building, Parks Road, Oxford, OX1 3QD
Tel: +44 1865-273883
Fax: +44 1865-273839


From: Olaf Storaasli <>
Date: 17 Mar 97 20:19:11 -0500
Subject: NASA Large-scale Analysis Symposium

NASA Langley is sponsoring the 4th National Symposium on
Large-Scale Applications on High-Performance Computers and
Workstations Oct 15-17, 1997 in Williamsburg VA. Proposed
paper abstracts are due April 15, 1997. More details (including
Call for Papers) is available on:

Kindly help us (NASA) publicize our Symposium by including notifying
your frioends and colleagues and Post the Call for Papers!

Sincerely yours,

Dr. Olaf O. Storaasli 757-864-2927


From: Stephen Watt <>
Date: Wed, 19 Mar 1997 18:43:57 +0100
Subject: Journal of Symbolic Computation Special Issue

Journal of Symbolic Computation
special issue on
Symbolic-Numeric Algebra for Polynomials
Extension of deadlines

The call for papers for this special issue of the Journal of Symbolic
Computation was announced late in 1996 through various channels. The
special issue is to bring together papers on the solution of algebraic
problems which utilize ideas and algorithmic techniques from numeric
computation together with standard and possibly extended procedures from
classical computer algebra.

The manuscripts should contribute to the understanding of the interaction
of symbolic and numeric computing and their results should be relevant for
the design of algorithms and their computer implementation. The full details
of the call may be found in ACM SIGSAM Bulletin 30 (3), 1996 (issue 117) or

While the response to this initiative has been encouraging, several
prospective contributors have asked for an extension of the submission
deadline. Others have considered submitting work continuing that of
their ISSAC 97 submissions, which requires some separation of intermediate

In any case, it has appeared that some significant contributions for
the special issue would not be possible with the initial submission
deadline of March 31, 1997. In order to provide equal opportunities
for all authors, we have modified the schedule for the JSC special issue
on Symbolic-Numeric Algebra for Polynomials as follows:

Deadline for submission of full papers: 15 May, 1997
Notification of acceptance/revision/rejection: 1 September, 1997
Final revised manuscripts due: 30 November, 1997
Appearance of special issue: 1998

Manuscripts should be submitted to either one of the two guest-editors who
will handle the preparation of this special issue:

Stephen Watt Hans J. Stetter
IBM T.J. Watson Research Center Technical University (115.2)
P.O. Box 218, Yorktown Heights, NY 10598 USA A-1040 Vienna, Austria


From: Bob McLatchie <>
Date: Thu, 20 Mar 1997 16:43:56 +0000
Subject: Practical Parallel Programming, a One Day Course

TWO chances to join in on this One Day Event

On Wednesday 16 April in London
OR Thursday 24 April in Edinburgh

Parallel Programming with BSPlib Boot-Up Day

Would you like to be able to write architecture-independent but
efficient parallel code in Fortran or C? You've tried MPI and PVM, and
would like a much simpler alternative that is just as effective, an
alternative that eliminates the risk of deadlock?

Or maybe you've heard of Bulk Synchronous Parallel (BSP) programming
available as part of the BSP Programming Environment being developed by
Oxford Parallel, and want to know more?

This thoroughly practical course provides instruction and a FREE copy of
the tools to take away with you. See
or E-mail or fax me for details.

Bob McLatchie Oxford Parallel
OUCL Wolfson Building Phone: +44 1865 273897
Parks Road Fax: +44 1865 273819
Oxford, England OX1 3QD e-mail:
University of Oxford Parallel Applications Centre


From: Tom Peacock <>
Date: Tue, 18 Mar 1997 15:40:36 +0000
Subject: Reduced Price Books from Oxford

Mathematics Sale
1st April - 31st May 1997

Oxford University Press is about to hold its first ever mathematics sale. The
sale applies throughout Europe and the Middle East and offers the mathematical
community a unique opportunity to buy recently published titles at large
discounts (up to 2/3 off). After 31st May all books will revert back to their
regular list prices.

How to get more information on titles featured in the sale:
Email with details of your address and OUP will send you a
catalogue by return of post. Also, full details of books and a list of
participating bookshops is available on the OUP web site

Availability of titles:
Please note: stocks of many titles are limited and will be sold strictly on a
"first come - first served" basis, so please order early to avoid

List of titles that may be of interest to the NA community follows, but there
are dozens of other books that may also be of interest.

Moving Finite Elements
M. J. Baines
0-19-853467-1 Hardback GBP47.50
Sale price GBP25.00

Lattice Methods for Multiple Integration
I. H. Sloan and S. Joe
0-19-853472-8 Hardback GBP41.95
Sale price GBP20.00

Best wishes
Tom Peacock, Oxford University Press, UK


From: Richard Brent <>
Date: Wed, 19 Mar 1997 14:06:32 +1100 (EST)
Subject: Research Position at Australian National University

The Computer Sciences Laboratory at the Australian National University
is advertising a two-three year Research Fellowship. The closing date
is 4 April 1997. Further information is available by anonymous ftp from
or by sending e-mail to


From: Ken Hinson <>
Date: Mon, 17 Mar 1997 14:41:12 -0500
Subject: Systems Administrator Position at Virginia Tech

Department of Mathematics
Virginia Tech

Manages a network of heterogeneous, multi-user UNIX servers and
workstations that provide a vital infrastructure for the Mathematics
Department's research, teaching, and outreach activities.

QUALIFICATIONS: Bachelors degree or equivalent training and experience in
computer science, mathematics or related field. Considerable experience in
system administration with a network of heterogeneous, multi-user UNIX
systems. Strong interpersonal and communication skills with experience in
supporting a diverse user community.

Full-time position. Normal Starting Pay: $39,940 SALARY IS NEGOTIABLE
based on qualifications. OPEN UNTIL FILLED.

For more information see

TO APPLY: Call (540) 231-5301 or (540) 231-6258 (TDD) or visit any Virginia
Employment Commission office to obtain a state application. The
application form is also available at

Return application (resume also suggested) to:
Personnel Services
First Floor
Southgate Center
Blacksburg, VA 24061-0318
Fax: (540) 231-3830

REFER to job number NA241S.

Individuals with disabilities desiring accommodation in the application
process should call for assistance. VIRGINIA TECH IS AN EO/AA EMPLOYER


From: Bob Mattheij <>
Date: Tue, 18 Mar 1997 12:23:34 +0100 (MET)
Subject: Research Position at Technical University of Eindhoven

Position in Scientific Computing

The Scientific Computing Group, department of Mathematics,
Technical University of Eindhoven, The Netherlands is inviting candidates
to apply for a Ph D position. This position is part of a larger cooperation
on glass research, jointly with the deoartments of chemical and mechanical
engineering and the Dutch glass industry.
The contrubutions from mathematics involve the development of
models for simulating viscous flows, like the pressing of glass
in a mould or the radiative heat exchange (see also article in
SIAM News 1996, 29/8, p 24).
The present project deals with numerical models for flow in a glass tank,
using finite volume discretisation, domain decomposition and
multigrid-like techniques.
The Scienttific Computing Group consists of some 20 people (10 Ph D
students), working on various problems, most of them related to industrial
applications. There is a close cooperation with computer
science groups (parallel computing, computer graphics), with
various engineering departments (in a joint graduate school)
as well as with industrial partners. The group is a node in the European
Consortium for Mathematics in Industry.

We are looking for a person with a thorough background in numerical
analysis and interest in mathematical modelling and numerical simulation.
The position is offered for a 4 year period; the initial monthly salary
is DFL 2114, increasing to DFL 3775 in the fourth year.
Prospective candidates should send their CV and proof of their qualifications
to the address below.

Further inquiries can also be made by fax or e-mail.

Prof Dr R.M.M. Mattheij Tel: +31 402 472080 (work)
Dept of Mathematics +31 492 536904 (home)
Technische Universtiteit Eindhoven +31 402 472753 (secretary)
PO Box 513 Fax: +31 402 442489
5600 MB Eindhoven The Netherlands e-mail:


From: Karin Remington <>
Date: Wed, 19 Mar 1997 11:26:18 -0500
Subject: Contents, Transactions on Mathematical Software

Table of Contents
Volume 23, Number 1 (March 1997)

Ronald Cools, Dirk Laurie, and Luc Pluym
Algorithm 764: Cubpack++ -- A C++ Package for Automatic
Two-Dimensional Cubature

Paola Favati, Guiseppe Fiorentino, Grazia Lotti, and Francesco Romani
Local Error Estimates and Regularity Tests for the Implementation of
Double Adaptive Quadrature

L. Machiels and M. O. Deville
Fortran 90: An Entry to Object-Oriented Programming for
Solution of Partial Differential Equations

Are Magnus Bruaset and Hans Petter Langtangen
Object-Oriented Design of Preconditioned Iterative Methods in Diffpack

Ali Bouaricha
Algorithm 765: STENMIN -- A Software Package for Large, Sparse
Unconstrained Optimization Using Tensor Methods

S. Cabay, A. R. Jones ,and G. Labahn
Algorithm 766: Experiments with a Weakly Stable Algorithm for
Computing Pade and Simultaneous Pade Approximants

A. J. Geurts and C. Praagman
Algorithm 767: A Fortran 77 Package for Column Reduction of
Polynomial Matrices


From: SIAM <>
Date: Thu, 20 Mar 97 14:30:14 EST
Subject: Contents, SIAM Control and Optimization

SIAM JOURNAL ON Control and Optimization
MAY 1997 Volume 35, Number 3


Asymptotically Efficient Adaptive Choice of Control Laws in Controlled Markov
Todd L. Graves and Tze Leung Lai

Block Triangular Decoupling for Linear Systems over Principal Ideal Domains
Naoharu Ito and Hiroshi Inaba

Configuration Controllability of Simple Mechanical Control Systems
Andrew D. Lewis and Richard M. Murray

Weighted Sensitivity Minimization for Causal, Linear, Discrete Time-Varying
Michel Verhaegen

Output-Induced Subspaces, Invariant Directions, and Interpolation in Linear
Discrete-Time Stochastic Systems
Anders Lindquist and Gyorgy Michaletzky

On the Puiseux Series Expansion of the Limit Discount Equation of Stochastic
W. W. Szczechla, S. A. Connell, J. A. Filar, and O. J. Vrieze

Constrained Regular LQ-Control Problems
G. Stefani and P. Zezza

Optimal Control for Holonomic and Nonholonomic Mechanical Systems with Symmetry
and Lagrangian Reduction
Wang-Sang Koon and Jerrold E. Marsden

Investigation of the Degeneracy Phenomenon of the Maximum Principle for Optimal
Control Problems with State Constraints
Aram V. Arutyunov and Sergei M. Aseev

Generalized Controlled Invariance for Nonlinear Systems
H. J. C. Huijberts, C. H. Moog, and R. Andiarti

Finite-Dimensional Filters. Part I: The Wei-Norman Technique
M. Cohen de Lara

Finite-Dimensional Filters. Part II: Invariance Group Techniques
M. Cohen de Lara

Optimization of Observations: A Stochastic Control Approach
Boris M. Miller and Wolfgang J. Runggaldier

State Maps for Linear Systems
Paolo Rapisarda and J. C. Willems


End of NA Digest