NA Digest Sunday, September 15, 1991 Volume 91 : Issue 37

Today's Editor: Cleve Moler

Today's Topics:

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: Panos Pardalos <pardalos@math.ufl.edu>
Date: Sun, 8 Sep 91 17:11:49 EDT
Subject: New Address for P.M. Pardalos

New address for P.M. Pardalos:

Panos M. Pardalos
303 Weil Hall
Department of Industrial and Systems Engineering
University of Florida
Gainesville, FL 32611

phone: 904-392-9011 office, 378-0281 home,
pardalos@math.ufl.edu pardalos@sun2.ise.ufl.edu


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

From: Andrew J. Cleary <ajclear@cs.sandia.gov>
Date: Thu, 12 Sep 91 11:41:05 MDT
Subject: Change of Address for Andy Cleary

Friends and colleagues,

As of September 13, 1991, I will be ending my position as postdoctoral
researcher at Sandia National Laboratories. After a brief hiatus, I
will be taking a new position as a research fellow at the Centre for
Information Science Research at the Australian National University.

My address there will be

Centre for Information Science Research
Australian National University
GPO Box 4
Canberra ACT 2601
Australia

My e-mail address at Sandia should stay active for a while.
My address at ANU will be

acleary@thrain.anu.edu.aus

Of course, my na-net address will be kept current, so that

na.cleary@na-net.ornl.gov

should be properly forwarded throughout the transition period.


Andy Cleary


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

From: Kristjan Jonasson <kj@numisun.ni.dth.dk>
Date: Mon, 9 Sep 91 14:11:50 MET DST
Subject: Sparse LP Software Sought

I would like to find software (preferably FORTRAN subroutines) for
solving large sparse linear programming problems. Ny intention is
to use these in a research project on nonlinear minimax problems.

Can anyone help me?

Kristjan Jonasson
Institute for Numerical Analysis
The Technical University of Denmark


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

From: Patricia Lamm <lamm@math.unc.edu>
Date: Mon, 9 Sep 91 15:40:57 EDT
Subject: Collocation of 1st Kind Integral Equations


I am looking for references containing convergence/nonconvergence
results for the collocation of certain first kind integral equations
when there is no numerical or data error present. In particular, I am
interested in the case where the kernel k(x,y) is not of the type in
which a transformation to a second-kind problem is possible (such as
occurs, for example, when the kernel or its derivative(s) is nonzero at
x=y). I need this information for an integral equation which is similar
to that which arises in an inverse heat conduction problem (i.e.,given
interior measurements, find the heat source).

Patricia K. Lamm
Department of Mathematics
Michigan State University
E. Lansing, MI 48824-1027
lamm@mth.msu.edu



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

From: Mohsen Maesumi <maesumi@cs4.lamar.edu>
Date: Mon, 9 Sep 91 17:48:16 -0400
Message-Id: <9109092148.AA29935@cs4.lamar.edu>
To: na.digest@surfer.EPM.ORNL.GOV
Status: R

Subject: FEM Information Requested

I'll appreciate receiving any pointers to
a) papers on the history of Finite Element Methods (FEM)
b) texts (preferably with floppy disk)
on FEM suitable for undergraduate students in math
c) public domain FEM packages (x-t, xy-t or xyz-t)

Thanks
Mohsen Maesumi
maesumi@cs4.lamar.edu


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

From: Stefan P. Mueller <ONM010%DE0HRZ1A.BITNET@YALEVM.YCC.Yale.Edu>
Date: Thu, 12 Sep 91 12:14:53 MEZ
Subject: Non-linear Least Squares Package in C

We are looking for a numerically reliable C-implementation of a fast
non-linear least squares algorithm. Our models will arise mainly from
compartmental modelling of tracer kinetic data in positron emission
tomography. Our current models have relatively few parameters and
allow the analytical specification of derivatives, however, we expect
situations in the future where this will not be possible.

For several years we have been using sucessfully the nl2sol package
for non-linear least squares fitting of a different class of models in
medical imaging [Dennis JE, Gay DM, Welsch RE. An adaptive nonlinear
least-squares algorithm. ACM Trans Math Software 1981;7:348-368]
(available from NETLIB, most recent incarnation is /port/dn[s2][fgp][b ]).
Since we are now moving our work to a SUN workstation where we do not
have a FORTRAN compiler available, we would like to know:

1. What is the current "state of the art" in non-linear least
squares? Has progress over the past 10 years rendered the DGW
algorithm obsolete as a general purpose non-linear least squares
package?
2. Is there a numerically reliable C-implementation of the current
"winner" available (hopefully trough the INTERNET)?

Thank you very much for your help! Since we do not have a numerical
analyst available, your pointers will be invaluable for our work.

Stefan


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

From: Jerzy Wasniewski <unijw@uts.uni-c.dk>
Date: Sun, 15 Sep 91 15:04:06 CET
Subject: Software for Sparse Linear Programming

Hello,

Does anyone have, or know where I could get, any public domain
software for Sparse Linear Programming Problem. We need this
software for the education purpose.
Thank you in advance.

Best regards,
Jerzy

Jerzy Wasniewski Email: jerzy.wasniewski@uni-c.dk
The Danish Computer Centre for Phone: + 45 42 883 999-2426
Research and Education (UNI-C), Fax: + 45 45 930 220
DTH, Bldg. 305,
2800 Lyngby, Denmark


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

From: George Miel <miel@nevada.edu>
Date: Wed, 11 Sep 91 16:41:40 -0700
Subject: Fingering, Gambling, and Pythagoreas

Contrary to what might be expected from the subject line,
this note does not explore relational meanings among
the listed topics. It simply addresses disparate
goals in one shot:
* For interested parties, particulars of my new
professional abode are:
George Miel, Dept. of Mathematical Sciences,
University of Nevada, Las Vegas, NV 89154
phone: 702-597-4447 fax: 702-597-4343
e-mail: miel@nevada.edu
* In this regard, I have no objection to being
"fingered" (Cf. last week's submission from Gene Golub)
as long as strangers don't find certain data, eg,
numbers of my home phone or credit cards. The current
Scientific American (pp. 154-164) describes intersting
facets of law and liberty in cyberspace.
* About cards, if you like counting them or related
activities, come and visit us in Lost Wages. Also
permissible are interest and discussion on environmental
modeling at the new Supercomputer Center in the Hughes
College of Engineering.
* Lastly, I am responding to the individual who requested
a pointer to the Moler-Morrison algorithm for Pythagorean
sums. I have lost your name and e-address. Hoping that
you read the na.digest, here are the original refs:
pp. 577-581 IBM j. Res. Develop. (27) 1983 and pp. 582-589 ibid.


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

From: Harry Yserentant <harry@tue-num2.mathematik.uni-tuebingen.de>
Date: Thu, 12 Sep 91 15:15:50 +0200
Subject: Position at the Universitaet Tuebingen (Germany)

The Universitaet Tuebingen is inviting applications for the tenured
position of a

Professor in Numerical Mathematics

at a level corresponding to an associate professorship. The text of
the official advertisement is given below.


EBERHARD-KARLS-UNIVERSITAET TUEBINGEN

An der Mathematischen Fakultaet der Universitaet Tuebingen ist im
Arbeitsbereich Numerische Mathematik des Mathematischen Instituts
eine

C3-Professur fuer Numerische Mathematik

zu besetzen. Gedacht ist in erster Linie an Bewerberinnen und Be-
werber, die in der Forschung ein anwendungsnahes, aktuelles
Gebiet wie etwa die Numerik der partiellen Differentialgleichungen
oder grosser Systeme als Schwerpunkt haben und bereit sind, mit
den in der Fakultaet vertretenen Arbeitsgruppen zu kooperieren.
In der Lehre wird eine regelmaessige Beteiligung an der Numerik-
ausbildung in den Studiengaengen der Fakultaet sowie anderer
Fakultaeten der Universitaet erwartet.

Einstellungsvoraussetzungen sind die Habilitation oder gleich-
wertige wissenschaftliche Leistungen.

Die Universitaet Tuebingen strebt eine Erhoehung des Anteils von
Frauen in Forschung und Lehre an und bittet deshalb qualifizierte
Wissenschaftlerinnen nachdruecklich um ihre Bewerbung.

Bewerbungen mit den ueblichen Unterlagen werden bis zum 31.Oktober
1991 erbeten an den Dekan der Mathematischen Fakultaet, Universitaet
Tuebingen, Auf der Morgenstelle 10, 7400 Tuebingen 1.


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

From: Carnegie Mellon <me@black.math.cmu.edu>
Date: Thu, 12 Sep 91 12:20:44 PDT
Subject: Position at Carnegie Mellon

The Department of Mathematics, Carnegie Mellon University, invites
applications for a senior level appointment in Computational Mathematics
/Numerical Analysis. We are particularly interested in candidates who will
enhance existing computational and analytical programs. These involve
continuum models in fluid dynamics and the mechanics of solids including
microstructure and phase transitions as well as other aspects of materials
science. Applicants should send a vita and list of publications and arrange
to have at least three letters of recommendation sent to: Chairman,
Computational Mathematics Search Committee, Department of Mathematics,
Carnegie Mellon University, Pittsburgh, PA 15213.

Carnegie Mellon University is an Affirmative Action/Equal Opportunity Employer.


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

From: Alessandro Russo <ALEX17%IPVIAN.BITNET@UICVM.uic.edu>
Date: Thu, 12 Sep 91 18:20:59 ITA
Subject: 6th Domain Decomposition Conference

Sixth International Conference on
DOMAIN DECOMPOSITION METHODS IN SCIENCE AND ENGINEERING
Como, Italy, June 15-19 1992

Organizing Committee

Chair:
Alfio Quarteroni, Politecnico di Milano and University of Minnesota

Donatella Marini, Universita' di Genova and I.A.N.- C.N.R., Pavia
Alberto Valli, Universita' di Trento, Povo (TN)
Alessandro Russo, I.A.N.- C.N.R., Pavia


Scientific Committee

J.H. Bramble, Cornell University, Ithaca
T.F. Chan, University of California, Los Angeles
R. Glowinski, University of Houston
G.H. Golub, Stanford University
Yu.A. Kuznetsov, USSR Academy of Sciences, Moscow
J. Periaux, Avions M. Dassault, Saint Cloud, France
A. Quarteroni, Politecnico di Milano and University of Minnesota
O.B. Widlund, Courant Institute of Mathematical Sciences, New York

Domain Decomposition is an increasingly important part of scientific
computing. Algorithms based on this approach have attracted the
attention of many numerical analysts and computational scientists.
Parallel computing systems are rapidly becoming more powerful and they
already offer quite cost effective alternatives to supercomputers for
solving many problems in the natural and engineering sciences.
Domain Decomposition methods, which reduce the solution of a problem
to a sequence of independent subproblems restricted to many individual
subregions of the given region, appear to offer the best promise for
the effective use of this powerful new technology.
The development of the field is not only driven by technological
change; there is also a very active development of more mathematical
aspects. This work is also shedding new light on some more traditional
research problems, e.g., multi-grid and classical iterative methods,
going beyond the traditional application of domain decomposition
methods to the numerical approximation of partial differential
equations.
A series of international Symposia have been held since 1987 and the
Proceedings, published by SIAM, have helped advance the field. The
interchange between the engineering and academic communities has
actively been encouraged and this Conference will also make every
effort to further foster connections between applied scientists,
computer scientists and numerical analysts.


Conference Themes

1. Numerical Analysis of Domain Decomposition
2. Block and Substructuring Methods
3. Multilevel Methods
4. Domain Decomposition for Time Dependent Problems
5. Interface Conditions for Heterogeneous Domain Decomposition
6. Parallel Implementation
7. Multibody Dynamics
8. Decomposition Methods in Microelectronics
9. Environmental Sciences
10. Applications in Science and Engineering


Invited Speakers

A temptative list of invited speakers includes:

V.I. Agoshkov, Department of Numerical Mathematics, Moscow
R. Bank, University of California at San Diego
J.H. Bramble, Cornell University, Ithaca
A.E. Brandt, Weizmann Institute of Science, Rehovot, Israel
C. Canuto, Politecnico di Torino, Italy
T.F. Chan, University of California at Los Angeles
M. Dryja, Warsaw University, Poland
T. Dupont, University of Chicago
A. Ecer, Purdue University, Indianapolis
C. Farhat, University of Colorado at Boulder
P. Fischer, Brown University, Providence
R. Glowinski, University of Houston
G.H. Golub, Stanford University
W. Gropp, Argonne National Laboratory, Argonne
D. Keyes, Yale University, New Haven
Yu.A. Kuznetsov, Department of Numerical Mathematics, Moscow
U. Langer, Technical University of Chemnitz, Germany
P. Le Tallec, Universite' de Paris Dauphine, France
P. Leca, ONERA, Chatillon, France
D. Marini, Universita' di Genova and I.A.N.- C.N.R., Pavia
S.V. Nepomnyaschikh, USSR Academy of Sciences, Novosibirsk
J.E. Pasciak, Brookhaven National Laboratory, Upton, Long Island
J. Periaux, Avions Marcel Dassault, Saint Cloud, France
R. Plemmons, Wake Forest University, Winston-Salem
B. Smith, Argonne National Laboratory, Argonne
J. Sun, Computing Center, Academia Sinica, Beijing, China
T. Tezduyar, Minnesota Supercomputer Institute, Minneapolis
O.B. Widlund, Courant Institute of Mathematical Sciences, New York
J. Xu, Penn State University, University Park
D.P. Young, The Boeing Company, Seattle


Location

The conference center is Villa Olmo, a beautiful neoclassic building
of the 18th Century facing the lake, located in Como, a city between
Milan and the Swiss border.
Como is served by the European railways and motorways system, and by
the nearby airports of Milan and Zurich.


Organization

The Conference is jointly organized by:

CNR, Istituto di Analisi Numerica, Pavia
Politecnico di Milano, Dipartimento di Matematica, Milano
Universita' di Trento, Dipartimento di Matematica, Povo (TN)


Format

The Conference will host about 30 invited lectures, as well as about
30 contributed presentations and a poster session.


Call for Contributions

Extended abstracts of a maximum of 1000 words on topics related to the
Conference are invited by December 1st 1991. Notification of
acceptance will be given by January 31st, 1992, at which stage
reccomendation concerning the format of the papers will be sent to
authors.
The Conference Proceedings will be published by S.I.A.M.,
Philadelphia.
Short abstracts of a maximum of 200 words are invited for poster
exhibition by December 1st 1991.

For more information, including a registration form, contact:

Sixth Domain Decomposition Conference
I.A.N.- C.N.R.
Palazzo dell'Universita'
Corso Carlo Alberto, 5
27100 Pavia, ITALY

Fax: +39-382-28079
e-mail: DOMDEC6@IPVIAN.BITNET


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

From: SIAM Publications Department <SIAMPUBS@WILMA.WHARTON.UPENN.EDU>
Date: Wed, 11 Sep 91 11:49 EDT
Subject: Contents, SIAM J. Control and Optimization

SIAM Journal on Control and Optimization
Vol. 30, No. 1, Janaury 1992

Lagrange Multipliers in Stochastic Programming
Sjur D. Flam

Algebraic Geometric Aspects of Feedback Stabilization
Shiva Shankar and V. R. Sule

Unbounded Solutions to the Linear Quadratic Control Problem
G. Da Prato and M. C. Delfour

Local Convergences and Optimal Shape Design
Wenbin Liu and J. E. Rubio

Stable Solutions of Real Algebraic Matrix Riccati Equations
Andre C. M. Ran and Leiba Rodman

Sufficiency Criteria via Focal Points and via Couples Points
Vera Zeidan

Properties of Energy-Minimizing Segmentations
Jayant Shah

The Revenge of the Linear Search Problem
Anatole Beck and Micah Beck

Hinf-Control by State-Feedback for Plants with Zeros on the Imaginary Axis
Carsten Scherer

Hinf-Optimization Without Assumptions on Finite or Infinite Zeros
Carsten Scherer

Boundary Velocity Control of Incompressible Flow with an Application to
Viscsous Drag Reduction
Max D. Gunzberger, Lisheng Hou, and Thomas Svobodny

The Discrete Time Hinf-Control Problem with Measurement Feedback
A. A. Stoorvogel

New Results in Pole Assignment by Real Output Feedback
Joachim Rosenthal

Information and Strategies in Dynamic Gams
P. Bernhard

On the Modelling and Exact Controllability of Networks of Vibrating Strings
E. J. P. George Schmidt


For more information contact Vickie Kearn, Publisher, SIAM, 3600 University
City Science Center, Philadelphia, PA 19104-2688. Phone: (215) 382-9800,
e-mail: siampubs@wharton.upenn.edu


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

End of NA Digest

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