NA Digest Wednesday, June 21, 1995 Volume 95 : Issue 25

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: Gene Golub <golub@sccm.Stanford.EDU>
Date: Wed, 14 Jun 95 0:42:56 PDT
Subject: Programs in Scientific Computing

A few weeks ago, I wrote:

> After talking to many students, I have the impression that it is hard
> for students to know where there are graduate programs in Scientific
> Computing and Applied Mathematics. If you have such a program and if
> you have a page on the WEB that is accessible, please send me a note
> and I will make a page available which will allow students (and
> others) to get a better perspective on our subject.

It turns out that there is already a WEB page with the desired information.

It's very easy to include your organization, too! I'm always amazed
what you can find in netlib.



From: Marc Prevost <>
Date: Wed, 21 Jun 1995 08:48:35 --100
Subject: Roger Apery

Roger Ap\`ery.

Professor Roger Ap\`ery , a prominent figure of the University of
CAEN (France),passed away after a prolonged illness on December 1994.
Roger Ap\`ery was born in ROUEN in 1916 of a mother of Flemisch
origin and a Greek father who had volunteered to serve in the French army
in 1914 in order to obtain French nationality.
After brilliant studies at the "Lycee Louis Le Grand" where he
distinguished himself several times in the "Concours general" , Roger Ap\`ery
was placed second among entrants at the ENS in 1936 and came first at the
"Agregation de Mathematiques".
Called up for the army in 1939 and a prisoner of war in June 1940 ,
he was released in October 1941 for health reasons. Appointed assistant
lecturer at the Sorbonne in 1942 , he joined a group of ENS students in the
French Resistance and became the leader of the National Front at the ENS.
In 1947 he defended his thesis in algebraic geometry " a l'italienne"
under the supervision of Paul Dubreil and was appointed Lecturer at Rennes
(the youngest ever in France). From 1949 until he retired in 1986 , Roger
Ap\`ery was a Professor at the University of CAEN where he created a research
team on algebra and number theory.
At the end of his career in 1977 , he made a sensational discovery
which was to make his name famous throughout the world. His proof of the
irrationality of the sum of the inverse of the cubes of integers by an
exceptionally clever method worthier of his Greek ancestors than of Bourbaki,
made him a legend.
In addition to a keen sense of provocation,Roger Ap\`ery enjoyed
playing the piano - his mother had taught him , chess,philosophy and ...
politics. Having joined the Camille Pelletan Radical Party at a young age
after the riots of 1934 , he resigned after Munich.Then , at the end of the
war,he once again became an active party member with Pierre Mendes-France.
As president of the Calvados Radical Party in the 60's he remained
active in politics until May 68. Being opposed to the reforms instituted
after 68 by Edgar Faure , he abandoned political life when he realized
University life was running against the tradition he had always upheld.

Many researchers have worked with the so-called Ap\`ery sequences to
rediscover his proof of the irrationality of Zeta(3)( See H.Cohen,A. Van Den
Poorten, E.Reyssat,F.Beukers,M.Prevost), to generalize his recurrence
relation in connexion with Numerical Analysis and Orthogonal Polynomials
( R.Askey,J.A. Wilson,A.L.Schmidt) or to study the congruence properties
of Ap\`ery numbers ( P.T.Young,Y.Mimura,I.Gessel,S.Chowla).
(From a text by Y. Hellegouarch).


From: Fred J. Hickernell <>
Date: Mon, 19 Jun 1995 21:38:55 +0800 (EAT)
Subject: An Electronic Abstract Server?

Has anyone considered setting up a listserver that would broadcast
abstracts of new results to interested recipients? One would subscribe by
giving one's name, e-mail address and the AMS classification codes
corresponding to one's areas of interest. When an abstract was received
by the server, it would be broadcast to those on the mailing list with the
corresponding classification codes. The abstract could also give the
location of the full preprint for downloading. Does such a server already
exist? If not, is there interst in setting one up? Would the na-net,
SIAM, AMS or some other group consider it? I would be interested to hear
others' opinions.

Fred J. Hickernell Email:
Department of Mathematics Office: (852) 2339-7015
Hong Kong Baptist University Fax: (852) 2336-1505
Kowloon Tong, HONG KONG Home: (852) 2609-1816


From: C Zhu <>
Date: Wed, 14 Jun 95 13:47:17 EDT
Subject: New Journal on Computation and Computational Mathematics


This is a serial publishing surveys with some original content on
mathematical aspects of ``computation'', including numerical computation.

The Editor-in-chief is

Charles Dunham,
Computer Science Department,
The University of Western Ontario
London, CANADA N6A 5B7

Members of the editorial board with numerical interests are

Lee Keener, University of Northern British Columbia

Alan Law, Memorial University of Newfoundland

We are looking for volume editors who would take responsibility for producing
a 6-12 article volume on a special topic of their choice. There is presently
room for individual contributions. In both cases submit a proposal to the
editor or write for details. The publisher is ABLEX, Norwood, New Jersey.


From: Olivier Coulaud <>
Date: Thu, 15 Jun 1995 18:45:39 +0200
Subject: Para++ Bindings for Message Passing Libraries

Para++: C++ Bindings for Message Passing Libraries

Para++ is a project supported by INRIA-Lorraine (FRANCE). Its aim is to
provide a user-level C++ interface to message passing libraries.

This is the first release (June 1995). There is a user guide which gives
detailed installation instructions and a quick tour of Para++.
An "installation" section of that guide gives the minimal instructions for
installation. (This guide can be found in Para++/doc/

To install Para++ :

Get para++-v0.9.tar.Z by anonymous ftp from
in the directory /pub/loria/numath/

zcat para++-v0.9.tar.Z | tar xvf -

cd Para++

Send bugs and comments to


From: Biswa Datta <>
Date: Tue, 13 Jun 95 16:49:56 CDT
Subject: Book by Datta on Numerical Linear Algebra

A few weeks ago ( Nadigest, April 9, 1995 ) , I announced the
publication of my book : Numerical Linear Algebra and
Applications, by Brooks/Cole publishing Company.
Unfortunately, the book was out of print that time.
I am pleased to report that the book has now been reprinted.

Here are the address and Telephone numbers of my publisher again.

Publisher : Brooks/ Cole Publishing Company
511 Forest Lodge Road
Pacific Grove, California 93950, USA
Tel : 1-800- 432-0563 ( for review )
1-800-354-9706 ( for purchasing )

Fax : (415) 592-3342 ( for review)
(606) 525-0978 ( for purchasing)

E-mail : ( for review)

If you are planning to use the book for your Fall or Spring
Semester classes , I urge you to obtain a copy from the second
print, even if you have a copy from the first print.

I look forward to receiving your kind feedback on the book.

Biswa Datta ( E-mail :


From: Trini Flores <>
Date: Tue, 13 Jun 95 09:43:11 EST
Subject: ACM-SIAM Symposium on Discrete Algorithms


Seventh Annual ACM-SIAM Symposium on Discrete Algorithms
(SODA), sponsored by ACM Special Interest Group on Algorithms and
Computation Theory and SIAM Activity Group on Discrete

January 28-30, 1996, Sheraton Colony Square Hotel
Atlanta, Georgia


This symposium concerns research on the use, design, and analysis
of efficient algorithms and data structures, in areas including,
but not limited to:

Combinatorial Structures; Computational Biology; Computational
Geometry; Databases and Information Retrieval; Discrete
Optimization; Graphs and Networks; Machine Learning; Number
Theory and Algebra; On-line Problems; Pattern Matching and Data
Compression; Random Structures; and Symbolic Computation.

The algorithms may be sequential, distributed, or parallel. They
may be analyzed either mathematically or by rigorous
computational experiments. In either case, the selection of
papers will be based on the extent to which the results yield new
insights for the design of efficient algorithms.

The program committee especially encourages submissions which
report on experimental and applied research. Special
consideration will be given to work that is motivated by
real-world problems. Experimental and applied papers are expected
to show convincingly that the algorithms or data structures
discussed are useful and efficient in a practical setting.

Program Committee

Eva Tardos (Chair), Cornell University; David Applegate, AT&T
Bell Laboratories; John Canny, University of California,
Berkeley; David Eppstein, University of California, Irvine;
Zvi Galil, Columbia University and Tel-Aviv University, Israel;
David R. Karger, Massachusetts Institute of Technology;
Anna R. Karlin, University of Washington; Nati Linial, Hebrew
University, Israel; Satish B. Rao, NEC Research Institute;
Jeffrey S. Vitter, Duke University; and Peter M. Winkler, AT&T
Bell Laboratories

Papers will be selected for presentation based on extended
abstracts. Authors wishing to submit a paper should send twelve
copies of an extended abstract (not a full paper) to:
SIAM Conference Coordinator, 3600 University City Science
Center, Philadelphia, PA 19104-2688, U.S.A.

Abstracts must be received at the SIAM office by JULY 14, 1995
(or postmarked by July 12, 1995 and sent airmail). This is
a firm deadline; submissions after the deadline will not be
considered. Letters of acceptance/rejection will be sent
by September 20, 1995.

Abstract Format. Abstracts should begin with the title of the
paper, each author's name, affiliation, and e-mail address,
followed by a succinct statement of the problems that are
considered in the paper, the main results achieved, an
explanation of the significance of the work, and a comparison to
past research. This material should be easily understood by
nonspecialists. Technical developments, directed toward the
specialist, should follow as appropriate. The entire extended
abstract must not exceed 10 pages (using 11 point or larger font,
with not less than one inch margins all around). Abstracts that
deviate significantly from these guidelines risk rejection
without consideration of their merits.

Electronic Access

Information regarding the Symposium can be accessed in electronic
format via SIAM's Gopher server: or through the
World Wide Web:

You can also contact SIAM by telephone: 215-382-9800; Fax: 215-
386-7999; E-Mail:


From: Wolfgang Mackens <>
Date: Wed, 14 Jun 1995 12:10:37 +0200 (MET DST)
Subject: Lothar Collatz Memorial Lecture and Dinner

Lothar Collatz Memorial Lecture and Dinner
Hamburg July 6, 1995, 6.30 p. m.
Bundesstr. 55
20146 Hamburg, Germany
"Geomatikum", Building of the Department of Mathematics
of the University of Hamburg
Lecture Theatre H2

The Department of Mathematics, the Institute of Applied Mathematics and the
Mathematical Society of Hamburg will organise a Colloquium on the eve of
July 6, 1995 in memory of Lothar Collatz, who would have celebrated his
85th birthday on this day. The colloquium will include a scientific lecture by

Karl Peter Hadeler (Tuebingen): "Reaktionstelegraphengleichungen"

and a gathering with a no host dinner will be offered afterwards. We cordially
invite you to participate in this event.

In case of your interest please contact us (by using the following e-mail address)
for further information. Expect to pay 36,- USD for the dinner.

Yours sincerely,

Gerhard Opfer

fax: +49 40 4123 5117


From: David F Griffiths <>
Date: Wed, 14 Jun 95 14:24:28 BST
Subject: Update on Dundee NA Conference


University of Dundee, Scotland, UK
27-30th June 1995

We currently have 208 registered participants and 121 submitted talks.

We can still take a limited number of late registrations.

For regulars to the conference, please note that the accommodation is
in a different hall of residence this year. Airlie Hall is situated
on the main campus. It is about a 5-10 minute walk from the rail station.

Sources of further Information

*** www users through the URL

through which it is also possible to register online. Details are
given of
speakers - titles and abstracts
full program
full list of participants
travel to and from Dundee (mainly trains from Edinburgh & London)
local map
candidates and timetable for the Leslie Fox Prize meeting

*** ftp users at the site

in the directory pub/na-conf.

*** Unix users, the command
will also provide a list of participants as well as a program.

The conference will be preceded on Monday, 26th June by talks from
candidates shortlisted for the Leslie Fox Prize.

Dr D F Griffiths
Biennial Conference on Numerical Analysis
Department of Mathematics and Computer Science
The University, Dundee, DD1 4HN
Scotland, UK

Tel: +44 (1382) 344467 e-mail:
FAX: +44 (1382) 345516 or:


From: C. Walsh <>
Date: Mon, 19 Jun 1995 15:21:58 +0100 (BST)
Subject: Postdoc Position at University Greenwich, London

Applications are invited for an EPSRC-funded Fellowship in the
Nano-Science Simulation Group to investigate a multi-level,
unified, modelling of fracture physics in adhesive joints. Such
an approach will involve a unified, computer-based, simulation of
materials properties at the nano, micro and continuum scales.
Both deterministic and stochastic Molecular Dynamics (MD)
simulation strategies will be employed to gain insights into the
initial formation and propagation of cracks at the atomistic
level. The micro- and macro-scale behaviour of these nano-scale
cracks will be studied via continuum theories of fracture
Applicants should have a strong basis in the application of
atomistic and molecular simulation methodologies in materials
science/condensed matter physics and be prepared to extend their
expertise into the continuum-scale modelling of material
The post is initially for a 2 year duration, starting from
September 1995, with the possibility of further extension at the
salary of 19000 Pounds per annum.
Applications, via e-mail and fax, enclosing the curriculum
vitae, the list of publications, names and e-mail and fax numbers
of three academic referees should be made to
Mr. Tex Dunstan
Research Office, Room 260, Island Site,
Woolwich Campus, Greenwich University
FAX: 0181-331-8695
Please quote Ref: HRT.
The University of Greenwich is an Equal Opportunities Employer


From: Philippe Langlois <>
Date: Mon, 12 Jun 1995 19:25:38 +0400
Subject: Postdoc Position at Universite de La Reunion, France.

Universite de La Reunion (La Reunion Island - France)
Mathematics and Computer Science Department

The Mathematics and Computer Science Department of the Universite de La Reunion
invites a postdoctoral candidate to apply for an ATER position (Temporary Attache
for Education and Research) for one year (may be renewed for a second) to
commence September 95.

* Education profile : teaching mathematics, applied mathematics or computer
Science at undergraduate or first year graduate level.

* Research profile : the candidate will work with the "Numerical
Quality of Scientific Software" group : computer arithmetic, methods and
tools for error-bounds in numerical algorithms, interval arithmetic algorithms,
self-validating numerical method, ...
Experience in this domain is absolutely necessary.

A PhD in applied or computational mathematics, or in computer science is
required (a "PhD-last year" could also apply). The monthly salary will be
approximately 15,000FF (~$3000).

The Universite de La Reunion is the french university in the Indian Ocean
(La Reunion Island is a french oversesas department). The Mathematics and Computer
Science department (~30 young teachers, researchers and PhD students) supports
an effective and up-to-date computational environment (3 SPARC servers 1000,
10 SUN SS10, 10 PM 7100, ...).

To apply, please submit by email : a curriculum vitae, a list of recent
publications, a statement of research interests and a letter of candidature.

Direct all material and questions to :

Closing date : 3rd July 1995

PS : Il est indispensable de parler correctement le francais : enseignement oblige !


From: Peter Kohl <>
Date: Fri, 16 Jun 1995 17:41:01 +0100
Subject: Postdoc and Research Positions at Oxford and Johns Hopkins


up to join teams of D. Noble (Oxford) and R. Winslow (Johns Hopkins) in 3D
modeling of electrical and mechanical properties of the heart. Ph.D. and strong
background in programming (ideally on parallel architecture computers)
required; basic knowledge in cardiac physiology of advantage. The work will
involve medium-term working visits to the overseas partner lab. Salary
commensurate with background and experience. Please send c.v. and names of
three referees to either Denis Noble, University Lab of Physiology, Parks Road,
Oxford OX1 3PT, UK, or Raimond L. Winslow, Dept Biomed Engineering, The Johns
Hopkins University School of Medicine, 720 Rutland Ave, Baltimore MD 21205, USA.
Job also advertised in Science, 16 June 1995


From: N. J. Ford <>
Date: Tue, 13 Jun 1995 09:18:38 +0100 (BST)
Subject: PhD Studentship at Chester College, UK

PhD Studentship at Chester College, UK

I have funding to support a PhD/MPhil student at Chester College
for two or three years from autumn 1995. The project is in
Computational Mathematics: Volterra Functional Equations, the
Numerical Solution of Evolutionary Problems.

The basis of the funding is that we are offering a bursary of
GB Pounds 5500 per year (approx) and offering to pay UK/EU student

The closing date for application is 7th July 1995. For details please
write to me

Dr Neville J Ford, Department of Mathematics, Chester College, Cheyney Road,
Chester, CH1 4BJ, UK

Tel -44-1244-375444 (ext 2392)
Fax -44-1244-373379


From: Heribert Blum <>
Date: Tue, 20 Jun 1995 12:27:25 +0200
Subject: Position at University Dortmund

Am Fachbereich Mathematik, der

ist eine


f=FCr Angewandte Mathematik / Numerik

Bewerberinnen und Bewerber sollen sich auf einem der Gebiete

wissenschaftlich ausgewiesen haben.

Erw=FCnscht sind Forschungsschwerpunkte in den Bereichen Signal-
oder Bildverarbeitung bzw. Optimale Steuerungen oder Parameter-
Es wird die Beteiligung an der Ausbildung der Studenten der
Mathematik, der Wirtschaftsmathematik und anderer Fachrichtungen,
sowie die Beteiligung an Aufgaben der Selbstverwaltung erwartet.
Habilitation f=FCr das Fach Mathematik oder habilitationsad=E4quate
Leistungen werden vorausgesetzt. Die Einstellungsvoraussetzungen
richten sich nach =A7 49 UG des Landes NRW.
Die Universit=E4t Dortmund strebt eine Erh=F6hung des Anteils von
=46rauen in Forschung und Lehre an und bittet deshalb Wissen-
schaftlerinnen nachdr=FCcklich um ihre Bewerbung. Die Bewerbung
geeigneter Schwerbehinderter ist erw=FCnscht.
Bewerbungen mit den =FCblichen Unterlagen werden erbeten bis
zum 20.07.1995 an den Dekan des Fachbereichs Mathematik
der Universit=E4t Dortmund, 44221 Dortmund.
Ausk=FCnfte sind erh=E4ltlich unter Telefon 0231 / 755(1) - 3050

Bewerbungen bitte nur per Post.


From: Stephan Mueller <>
Date: Tue, 20 Jun 1995 10:07:43 -0700
Subject: Position at Silvaco in Santa Clara, CA

Senior Development Engineer Position -
3-D Semiconductor Process/Device Simulation

Silvaco International is located in Santa Clara, CA. We are the
leading supplier of semiconductor technology CAD software worldwide.
For a new 3-D parallel process/device simulator project we currently
have an exciting new opening for a candidate with a strong NA

Duties would include the development, improvement or parallelization
of software in a broad variety of fields, for example sparse linear
solvers, nonlinear solvers, transient integration methods, and
continuation methods.

The candidate should have at least a four year technical degree
(MSc/PhD preferred). We require at least two years C programming
experience. Candidates with previous exposure to C++ are
preferred. Any experience with semiconductor device or process
simulation would be beneficial.

For more information about our company and products please have a look
at our WWW page:

Applications must be submitted by letter or email, outlining career
objectives and reasons for applying, and include a current resume and
appropriate references. Send your application to:
Dr. Stephan Mueller
SILVACO International
4701 Patrick Henry Drive #1
Santa Clara, CA 95054, USA


From: William Symes <>
Date: Tue, 20 Jun 1995 13:31:11 -0500 (CDT)
Subject: Position at Rice University

Rice University Computational & Applied Mathematics Department

The Department of Computational & Applied Mathematics invites applications
for tenure-track assistant professorship appointments in all computational
areas of mathematical programming, particularly integer programming,
combinatorial optimization, and interior point methods. Applicants should
demonstrate both breadth of interest and promise in research and teaching.
Appointment at more senior rank may be possible for exceptional candidates
with appropriate records of accomplishment.

Rice University is a private research university with an exceptional
undergraduate population and a long tradition of excellence in undergraduate
science and engineering education. The Computational & Applied Mathematics
Department hosts research programs in Linear and Integer Programming,
Nonlinear Optimization, Numerical Linear Algebra, Parallel Computing,
and Numerical Solution, Optimal Design, and Inverse Problems for Partial
Differential Equations. The department also has close ties to the Center
for Research in Parallel Computing at Rice, an NSF funded Science and
Technology Center.

Please furnish vita, transcripts, reprints, and three letters of
recommendation before 31 December 1995 to:

Chair, Staffing Committee
Computational & Applied Mathematics Department
Rice University
PO Box 1892
Houston, TX 77251-1892

Rice University is an affirmative action/equal opportunity employer.


From: Eamonn Cahill <>
Date: Wed, 21 Jun 1995 13:19:45 +0100
Subject: Position at Hitachi Dublin Laboratory

Hitachi Dublin Laboratory

The Hitachi Dublin Laboratory was founded in 1989 as part
of Hitachi's world-wide diversification of research. It
was created as a research laboratory within Trinity College,
University of Dublin. Basic and applied research is carried
out in the general area of Advanced Computing. Currently
there are thirteen employees. The laboratory is engaged
in several collaborative projects with various departments
of the university as well as with other European institutions
and Hitachi research laboratories.

Due to expansion, there is currently a vacancy for one
researcher in the parallel computing research group.
The successful candidate will work as part of a small
specialised team whose brief it is to implement key
message passing paradigms on Hitachi's parallel
architectures. Candidates should have a background in
computer science, mathematics or a numerate discipline.
A higher degree, preferably at doctorate level in either
parallel computing, networking or communications would
be advantageous. Highly developed skills in C and Unix,
particularly at a systems networking and communications
context are essential. In addition, some experience with
message passing software such as MPI and PVM is desirable.
The successful candidate will be based at one of the company's
development divisions in Japan for an initial period of
approximately one year. After this period the candidate
is expected to return to the Dublin Laboratory to join
the research group there.

Applicants should post, fax or email a Curriculum Vitae to:

Dr. M. Abe, Laboratory Manager, Hitachi Dublin Laboratory,
Trinity College, University of Dublin, Dublin 2, Ireland.

Fax: + 353 1 6798926.

The closing date for applicants is July 10th, 1995.


From: Lieke v.d. Eersten-Schultze <>
Date: Tue, 13 Jun 1995 13:31:19 +0200
Subject: Contents, Mathematics of Control, Signals, and Systems

Mathematics of Control, Signals, and Systems (MCSS)
Volume 7, Issue 4
Contributed by Jan H. van Schuppen (

G.E. Coxson and C.L. DeMarco,
The Computational Complexity of Approximating the Minimal Perturbation
Scaling To Achieve Instability in an Interval Matrix,
pp. 279-291.

V.V. Veeravalli, T. Basar, and H.V. Poor,
Decentralized Sequential Detection with Sensors Performing Sequential
pp. 292-305.

A.-J. van der Veen and P. Dewilde,
Embedding of Time-Varying Contractive Systems in Lossless Realizations,
pp. 306-330.

V. Solo,
On the Stability of Slowly Time-Varying Linear Systems,
pp. 306-350.

K.A. Morris,
State Feedback and Estimation of Well-Posed Systems,
pp. 351-388.

The new address for submissions is:

J.H. van Schuppen
Co-Editor MCSS
P.O. Box 94079
1090 GB Amsterdam
The Netherlands

E-mail inquires regarding submission should be addressed


End of NA Digest