NA Digest Friday, June 25, 1993 Volume 93 : Issue 23

Today's Editor:

Cleve Moler
The MathWorks, Inc.
moler@mathworks.com

Submissions for NA Digest:

Mail to na.digest@na-net.ornl.gov.

Information about NA-NET:

Mail to na.help@na-net.ornl.gov.

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



From: Cleve Moler <moler@mathworks.com>
Date: Fri Jun 25 22:23:51 EDT 1993
Subject: Hey! We're Back!


Hey! We're Back!


It's been almost four weeks since the last NA Net Digest. The first
half of the delay was because we lost "Surfer", the venerable Sequent
computer that we've used ever since NA Net moved to Oak Ridge.
The second half of the delay was because I've been travelling to
a few places where I didn't have convenient net access.


We may have lost some of the submissions during this period. So, if you
sent something in over the last month that isn't including in today's
issue, please sent it again. Same address:


na.digest@na-net.ornl.gov


NA Net, and its close associate, Netlib, are now running on a couple of
Sun SPARCs that are either at Oak Ridge or the University of Tennessee --
I'm not sure which. I log in to a machine with an Oak Ridge address,
but most of the work is being done by a crew at the University.
Thanks to Keith Moore and the rest of the crew for getting things
working again.


-- Cleve Moler




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


From: Nick Higham <higham@vtx.ma.man.ac.uk>
Date: Fri, 11 Jun 93 14:45:51 BST
Subject: NA Technical Reports from Manchester


Technical reports of the Manchester Centre for
Computational Mathematics (University of Manchester and UMIST) are now
available by anonymous ftp from


vtx.ma.man.ac.uk (130.88.16.2)


in pub/narep. The files README and index.bib explain how to access
the reports and the contents of the directory.
Not all recent reports are present, but more will be added in the near future.


Included among the reports is the Annual Report 1992 of the Manchester
Centre for Computational Mathematics, which is file
pub/narep/narep233.dvi.Z


---Nick Higham




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


From: Rob Bisseling <bisselin@math.ruu.nl>
Date: Mon, 14 Jun 93 11:51:17 +0200
Subject: Change of Address for Rob H. Bisseling


Change of address for Rob H. Bisseling


On June 1, 1993, I moved from the Koninklijke/Shell-Laboratorium, Amsterdam
to the Department of Mathematics, University of Utrecht, where I have been
appointed associate professor in Computational Science.


My new address is:


Rob H. Bisseling
University of Utrecht
Department of Mathematics
P.O. Box 80010
3508 TA Utrecht
The Netherlands


email: bisseling@math.ruu.nl
telephone: 030-531481
telefax: 030-518394




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


From: Yves V. Genin <Genin@auto.ucl.ac.be>
Date: Thu, 17 Jun 93 11:15:03 +0200
Subject: Change of Address for Yves Genin


Dear friends and colleagues,
Would you please note my change of address effective immediately:


Professor Yves V. Genin
U.C.L. Cesame
Batiment Euler
4-6, avenue George Lemaitre
B-1348 Louvain-La-Neuve (Belgium)
Tel: 32-10-478017
Fax: 32-10-472180
Email:genin@auto.ucl.ac.be




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


From: John Prentice <johnp@amber.unm.edu>
Date: Tue, 15 Jun 93 19:42:27 MDT
Subject: Information on Numerical Methods for Porous Flow Needed


It has been a few years since I was last doing any work in numerically
modeling flow in porous media. When I last was exposed to it, the
standard finite difference techniques being used were truly horrible, very
diffusive first order upwind differencing techniques that smeared
concentration fronts and often lead to spurious results. I am again
working in the field and I am anxious to come up to speed on what the
state of the art methods are. From what I can see, the same old
bad techniques are still pretty widely used in the remediation and
petroleum communities, but I assume that there are other techniques out
there that are better and just haven't made it into general use yet.


So, with that little diatribe out of my system :-) , I would much
appreciate pointers to papers discussing modern numerical methods for
flow in porous media. My particular interest is multi-phase flow
of chemical groundwater contaminants. I am also very interested in
work that incorporates a pressure equation of state into the
description of the flow.


Many thanks!


John


Dr. John K. Prentice
Quetzal Computational Associates
3200 Carlisle N.E., Albuquerque, NM 87110-1664 USA
Phone: 505-889-4543 Fax: 505-889-4598 E-mail: quetzal@aip.org




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


From: Hiroshi Murakami <hiroshi@teine.chem2.hokudai.ac.jp>
Date: Fri, 18 Jun 93 14:38:10 JST
Subject: Formulae Tables. (Bateman Manuscript Project)


Has anyone know whether if the book: "Higher Transcendental Function"
of Bateman Manuscript Project of CALTECH is still available from some
publisher today.
I found them (3 volumes of Higher Transcendental Function
and 2 volumes of the integral transformation) at my university library
of printed date in 1953 from McGraw-Hill. Are they out of print?
I would like to keep them for me at home.


( BTW. I wonder how much percentage the books covered the
Prof. Bateman's lifework collections? )


I wonder if the similar or larger formulae tables available by the
books or CD's, or netlib type file either Postscript or TeX style files.


I guess the copyright law prevents even the formulae tables stored by
the scanner or typed-in to re-distribute.


Thank you in advance.
Hiroshi Murakami
hiroshi@teine.chem2.hokudai.ac.jp




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


From: Phil Davis <AM188000@BROWNVM.brown.edu>
Date: Sun, 20 Jun 93 09:46:16 EDT
Subject: New Book by Phil Davis


Notice of a new publication containing some developments of interest to num
erical analysts.


"Spirals:From Theodorus to Chaos"
by Philip J. Davis


Publisher: AK Peters Ltd.
289 Linden St.
Wellesley, MA., 02181


This book contains supplements by Walter Gautschi of Purdue and Arieh
Iserles of Cambridge University relating to the summation of very slowly
convergent series and to a theoretical analysis of certain non-linear
systems of difference equations.




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


From: Michael Todd <miketodd@cs.cornell.edu>
Date: Wed, 23 Jun 93 09:54:54 -0400
Subject: Dantzig Prize: Call for Nominations


CALL FOR NOMINATIONS FOR THE
GEORGE B. DANTZIG PRIZE 1994


Nominations are solicited for the George B. Dantzig Prize,
administered jointly by the Mathematical Programming Society (MPS)
and the Society for Industrial and Applied Mathematics (SIAM).
This prize is awarded to one or more individuals for original
research which, by virtue of its originality,
breadth, and depth, is having a major impact on the field of mathematical
programming. The contributions eligible for consideration must be publicly
available and may address any aspect of mathematical programming in its
broadest sense. Strong preference is given to contributions by
individuals under 50 years of age.


The prize will be presented at the Mathematical Programming Society's triennial
symposium to be held August 15-19, 1994, in Ann Arbor, Michigan, USA. Past
Dantzig Prize recipients have been: M.J.D. Powell and
R.T. Rockafellar in 1982, E.L. Johnson and M.W. Padberg in 1985,
M.J. Todd in 1988, and M. Groetschel and A.S. Nemirovsky in 1991.


The Prize Committee members are: Michael J. Todd, Chair, Martin Groetschel,
Ellis L. Johnson, and R. Tyrrell Rockafellar.


Please send nominations to Michael J. Todd, School of Operations Research &
Industrial Engineering, 206 Engineering & Theory Center Building, Cornell
University, Ithaca, New York, 14853-3801, USA, or electronically to
miketodd@cs.cornell.edu. Nominations are due by September 30, 1993 and should
provide a brief one or two page description of the nominee's outstanding
contributions and, if possible, a current resume including a list of the
nominee's publications.


Michael J. Todd


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


From: John Pryce <pryce@rmcs.cranfield.ac.uk>
Date: Wed, 9 Jun 93 9:57 BST
Subject: Benin Conference on Scientific Computing, Nigeria


Benin Conference on Scientific Computing, Nigeria


25-29 January 1994 is the date of the 6th Benin International Conference on
Scientific Computing held at the University in Benin City, Nigeria, under
the direction of Professor Simeon Fatunla, Head of Mathematics.


This has become one of the premier scientific events in West Africa, which
the Nigerian Federal Government supports financially and by sending
Ministerial representatives to the Strategic Symposia which take place
alongside the technical mathematical side. It is regularly attended by
visiting speakers from Europe, N. America, Japan and elsewhere in the world.


The main topics this time will include:
Ordinary and partial differential equations; DAEs; computational
fluid dynamics; continuous optimization; automatic differentiation;
mathematical packages and numerical software; software engineering;
large scale computing.


We invite people from academia and industry with interests in these areas
to participate. You will need to make your own arrangements for travel to
Lagos (currently a January 1994 London - Lagos return is available for #475
sterling or less). The conference sees to your reception at Lagos airport,
transport between Lagos and Benin, and accommodation in hotel or University
guesthouse.


Why go? Your expertise contributes to a forum where a growing number of
well-trained young Nigerian mathematicians and computer scientists make
international contacts, learn about new developments and present their work
which often applies numerical computation to Nigerian industrial or
governmental problems. The conference is also great fun and includes trips
to places of note, a theatre production and other social events. The past
few conferences have had speakers from Canada, China, Germany, Holland,
Japan, Spain, UK, Canada and USA among others.


Any enquiries please to Dr John Pryce, conference co-director,
Software Engineering Group, RMCS,
Shrivenham, Swindon, SN6 8LA.
(e-mail: pryce@uk.ac.cranfield.rmcs; e-mail on na-net: na.pryce),


Professor Bob Russell at Simon Fraser University (na.brussell), who was
there in 1992, can also answer your questions.




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


From: Katalin Balla <H153Bal%HUELLA.BITNET@utkvm1.utk.edu>
Date: 18 Jun 93 14:53:38 +0100
Subject: Conference on Numerical Methods in Hungary


PRELIMINARY ANNOUNCEMENT


Continuing the traditions of the International Conferences on Numerical
Methods organized by the Bolyai Janos Society in Hungary (Keszthely:
1968, 1973, 1977, Miskolc: 1986, 1990), the 6th Conference will be held
on 22-26 August, 1994 in Miskolc, on the University campus.


As at the previous meetings, the SCIENTIFIC PROGRAM will be focused on
the
THEORY of NUMERICAL METHODS,
numerical algebra and
numerical solution of differential equations,
both ordinary and partial.
The theory of numerical optimization is included for the first time.
New numerical methods developed for the solution of problems in applied
sciences (in computational mechanics, fluid dynamics, geophysics, etc.)
may also be presented.


One-hour plenary LECTURES and half-hour lectures in sections will be the
principal part of the program. To present other results, short
communications in sections for each field of interest will be held and a
poster section will be organized. An opportunity will be given to the
participants who wish to show their public domain software in numerical
methods.


ORGANIZING COMMITTEE:
Chair: Rozsa, Pal (Technical University of Budapest)
Scientific Secretary: Balla Katalin (Computer and Automation Institute,
Hungarian Academy of Sciences)
Members: Stoyan, Gisbert (Eotvos Lorand University, Budapest)
Farago, Istvan (Eotvos Lorand University, Budapest)
Galantai, Aurel (Miskolc University, Miskolc)
Szarka, Zoltan (Miskolc University, Miskolc)


The first announcement and call for papers is scheduled for the end of
1993. If you are interested in attending the conference and wish to be
on the mailing list, SEND NOW, please, a MESSAGE to K. Balla,
h153bal@ella.hu (or na.balla@na-net.ornl.gov)




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


From: Aad van der Steen <actstea@cc.ruu.nl>
Date: Tue, 22 Jun 1993 17:16:18 +0200 (MET)
Subject: Fourth EuroBen Workshop


ANNOUNCEMENT OF THE FOURTH EuroBen WORKSHOP


The EuroBen Benchmarking Group has been founded in 1990. The group promotes
rationalisation and standardisation of benchmarking procedures for scientific
high-performance computing.


After three earlier successful workshops in Paris, Utrecht, and Regenburg the
EuroBen Benchmarking Group would like to draw your attention to the 4th EuroBen
Workshop. It will be held in Zurich, Switzerland, at 30 September--1 October,
1993.
Subjects that will be discussed there are:


-- The contents of the EuroBen benchmark modules.
-- Issue for a throughput benchmark and an interactive benchmark.
-- Discussion of recent benchmark results.
-- Standard (re)presentation of benchmark results.
-- Rigorous interpretation.
-- Further opportunities for cooperation with other benchmark groups in
Europe, Japan, and the USA.
-- Any other issue that could improve benchmarking of scientific
high-performance computers.


As in the earlier workshops, the upperbound for participation is 35--40 people
because of the desired interaction and discussions between participants. For
more information, participation, and/or proposals for contributions please
contact:


-- Dr. Armin Friedli
ETH-Zentrum, RZ
IPS Supercomputing
CH-8092 Zurich
Switzerland
Tel. +41-1-2563440 or 2565574
Fax. +41-1-2610468
Email: Friedli@ips.ethz.ch


or


-- Aad J. van der Steen
EuroBen
c/o Academic Computing Centre Utrecht
Budapestlaan 6
3584 CD Utrecht
The Netherlands
Tel. +31-30-531444
Fax. +31-30-531633
Email actstea@cc.ruu.nl




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


From: Tor Sorevik <tors@ii.uib.no>
Date: Wed, 9 Jun 1993 08:44:54 +0200
Subject: Position at University of Bergen, Norway


The Norwegian Supercomputing Board has awarded a grant to the Laboratory of
Parallel Processing (Para//ab) at Department of Informatics, University of
Bergen, for establishing a national center in parallel computing.
As a part of this project Para//ab will instal an Intel Paragon XP/S
supercomputer with 98 nodes.


To promote support and research on this computer, Para//ab will increase its
staff with the following positions:


SUPPORT STAFF
This person will join a group working with tasks related to operation and
maintenance of the supercomputer and user support, such as system
administration, system programming, diagnostic, installation and
maintenance of software, program development and training of users.
A master degree or equivalent is required.
Verification of skills and experience with parallel computers will be
required.
The position is vacant for a period of three years.


Salary is NOK 15.157 - 18.169 per month, depending on education and
previous job experiences.(USD 1 = NOK 6,90)


PH.D. SCHOLARSHIP
The applicant must hold a master degree (or equivalent) within informatics
or related area. The scholarship is for a period of three years.
The candidate is supposed to take part in the department's Ph.D. program
and the thesis will have to be within the subject of parallel computation.
The application should contain a proposed plan for accomplishment of the
dr.scient education.


Salary: NOK 16.924 per month


Women are particularly encouraged to apply. If the appointment committee
finds several candidates to be approximately equally qualified, the rules
contained in the Equal Opportunities Protocol for the University of Bergen
will be applied.


Further information can be obtained from professor Trond Steihaug
(email: trond@ii.uib.no) or researcher Tor Sorevik (email: tors@ii.uib.no)
at the Department of Informatics.


Application must contain a complete overview of education and professional
experience. Three sets of copies of certificates should be enclosed to
the application, which is to be sent to the Department of Informatics,
Hoyteknologisenteret, N-5020 Bergen, Norway, within July 1 1993.
Applicants for the scholarship must also enclose three copies of their
scientific publications and three copies of a list of these publications.




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


From: Bob Russell <russell@cs.sfu.ca>
Date: 16 Jun 93 15:21 -0700
Subject: Position at Simon Fraser University, Canada


Simon Fraser University


Applications are invited for a two-year Research Associate-
ship in the Department of Mathematics & Statistics, to com-
mence September 1, 1993, or as soon as possible thereafter.
The salary will be in the $25,000 to $30,000 range. Appli-
cants must have completed a PhD in Numerical Analysis. Fam-
iliarity with a unix environment, and prefereably with Sili-
con Graphics equipment, is necessary. The successful appli-
cant must have research experience in numerical PDEs, in
areas which specifically include adaptive grid methods,
spectral methods, and multigrid methods, so that they will
be involved at the outset in research projects with Profes-
sors R. D. Russell, T. Tang, and M. R. Trummer.
Applicants should send a curriculum vita and arrange for
three letters of recommendation to be sent by August 1, 1993
directly to
Professor Robert D. Russell
Department of Mathematics & Statistics
Simon Fraser University
Burnaby, BC V5A 1S6 Canada
FAX: (604)291-4947
E-mail: rdr@cs.sfu.ca
In accordance with Canadian immigration requirements, this
advertisement is directed to Canadian citizens and permanent
residents.




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


From: Brent Lindquist <lindquis@ams.sunysb.edu>
Date: Thu, 17 Jun 93 13:23:39 EDT
Subject: Position at SUNY, Stony Brook


POSTDOCTORAL POSITION IN APPLIED MATHEMATICS
UNIVERSITY AT STONY BROOK
Department of Applied Mathematics and Statistics


(This is a followup posting to an announcement of a posdoctoral position
which appeared in NA Digest, V. 92, # 41, Nov 1992.)


The Department of Applied Mathematics and Statistics at the University
at Stony Brook anticipates a postdoctoral position available for the
academic year of 1993/94. We are seeking qualified applicants to pursue
industrial applications of mathematics. Experience in such topics as
numerical analysis, modeling, or partial differential equations will be
prefered.


If you responded to our original announcement and are still interested,
please send e-mail to


lindquis@ams.sunysb.edu

and we will re-open your application file. If you are responding for
the first time, a vita, a brief description of research interests, and
at least three letters of recommendation should be sent to


Prof. Brent Lindquist
Department of Applied Mathematics and Statistics
University at Stony Brook
Stony Brook, NY 11794-3600.


Applications in the form of plain TeX or LaTeX files can be sent via
electronic mail to


lindquis@ams.sunysb.edu.


The University at Stony Brook is an Equal Opportunity Employer. We
especially solicit applications by women and minorities.




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


From: Bernard Tourancheau <btouranc@cs.utk.edu>
Date: Thu, 24 Jun 1993 14:45:59 -0400
Subject: Invited Computer Science Professor Position in France


IMMEDIATE OPENING POSITION IN COMPUTER SCIENCE AT THE ENS LYON (FRANCE)


A one year position at professor level will be open at the Ecole Normale
Superieure de Lyon (ENS Lyon), starting in September or October 1993.


The ENS Lyon is one of the most famous high schools in France. It has
status very similar to that of a Scientific University. Disciplines
studied at the ENS Lyon are Mathematics, Computer Science, Physics,
Chemistry, Biology and Geology. ENS Lyon hosts research laboratories
covering all scientific disciplines taught to its students. Students
enter the school 2 or 3 years after the baccalaureat. The program of
2nd year corresponds thus to the 4th academic year.


The teaching activities consist of 2 courses per semester. Main domains
of expertise are operating systems, languages and databases. It will be
done within the "Departement d'Informatique et de Mathematiques (DMI)"
which involves about 20 full or part time faculty members. The DMI
offers programs in Mathematics and Computer Science designed for
undergraduate students in their 3rd and 4th academic year, severely
selected after a nation-wide examination (1000 candidates each year).
Approximately 50 students are annually in these programs (30 in
Mathematics, 20 in Computer Science). Introductory courses are also
given to students in other disciplines (Physics, Biology). Most of the
ENS Lyon students eventually end up with a PhD.


The research activities should be in relation with parallel languages
and systems. The researcher will be associated with research currently
done at LIP, the Computer Science Laboratory of the ENS Lyon. LIP
research activities are entirely devoted to the domain of parallelism
from models to architecture. With 60 researchers (20 permanent
researchers and 40 PhD students) and a laboratory equipped with
state-of-the-art parallel computing equipment, it is one of the most
advanced centers for research in parallel processing in Europe.


A minimum of a PhD degree with several years of teaching and research
experience is needed.


Salary corresponds to an assistant professor position: FF 16 000 - FF 20 000
per month, depending on the experience.


Lyon is a well known international city and a great place to live, nearby
the Alpes mountains.


If you are interested, please immediately send a mail and
confirm before July 5th 1993, by sending a detailed
resume via FAX & via E-Mail or cover letter.


========================= Contact Info: ===============================
Jean-Louis Duclos, Tel: (33) 72-72-85-52
Public relation Email: jlduclos@lip.ens-lyon.fr
LIP/ ENS Lyon Fax: (33) 72-72-80-80
46, Allee d'Italie,
F 69364 Lyon Cedex 07




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


From: SIAM <wilshuen@siam.org>
Date: Wed, 09 Jun 93 13:49:30 EST
Subject: Contents, SIAM Optimization


SIAM JOURNAL ON OPTIMIZATION


August 1993, Volume 3, Number 3


CONTENTS


Nonsmooth Equations: Motivation and Algorithms
Jong-Shi Pang and Liqun Qi


A Newton Method for Convex Regression, Data Smoothing, and Quadratic
Programming with Bounded Constraints
Wu Li and John Swetits


Second-Order Multiplier Update Calculations for Optimal Control Problems
and Related Large Scale Nonlinear Programs
J. C. Dunn


A NonInterior Continuation Method for Quadratic and Linear Programming
Bintong Chen and Patrick T. Harker


An Implementation of the Dual Affine Scaling Algorithm for Minimum-Cost
Flow on Bipartite Uncapacitated Networks
Mauricio G.C. Resende and Geraldo Veiga


Convergence Analysis of a Proximal-like Minimization Algorithm Using
Bregman Functions
Gong Chen and Marc Teboulle


A Lagrangian Relaxation Algorithm for Multidimensional Assignment
Problems Arising from Multitarget Tracking
Aubrey B. Poore and Nenad Rijavec


Manifold Structure of the Karush-Kuhn-Tucker Stationary Solution Set
with Two Parameters
Ryuichi Hirabayashi, Masayuki Shida, and Susumu Shindoh


Numerical Experience with Limited-Memory Quasi-Newton and Truncated
Newton Methods
X. Zou, I. M. Navon, M. Berger, K. H. Phua, T. Schlick, and F. X. Le Dimet


A Globally Convergent Method For lp Problems
Yuying Li


A Collinear Scaling Interpretation of Karmarkar's Linear Programming
Algorithm
J. C. Lagarias


Automatic Column Scaling Strategies for Quasi-Newton Methods
Marucha Lalee and Jorge Nocedal


Multi-Objective Control-Structure Optimization Via Homotopy Methods
Joanna Rakowska, Raphael T. Haftka, and Layne T. Watson




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


From: SIAM <tate@siam.org>
Date: Tue, 08 Jun 93 14:30:29 EST
Subject: Contents, SIAM Review


SIAM Review
Table of Contents, Vol. 35, No. 3, September 1993


Articles


Mathematical Models of Running
W. G. Pritchard


Semi-infinite Programming: Thoery, Methods, and Applications
R. Hettich and K. O. Kortanek


Some Perspectives on the Eigenvalue Problem
David S. Watkins


Classroom Notes in Applied Mathematics
Problems and Solutions
Book Reviews
Chronicle




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


End of NA Digest

**************************
-------