NA Digest Sunday, February 28, 1999 Volume 99 : Issue 09

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: NA Digest <>
Date: Sun 02/28/1994 14:36:09 EST
Subject: NA Digest Calendar

The Netlib Conferences Database is on the Web at:

NA Digest Calendar
Date Topic Place NA Digest #

Mar. 15 Parallel Sparse Matrix Solvers Edinburgh, Scotland 06
Mar. 15-17 Scientific Computations Beirut, Lebanon 48
Mar. 16-18 BLAS Technical Forum Oak Ridge, TN 07
Mar. 22-24 Optimization Trier, Germany 03
Mar. 22-26 Numerical Analysis Workshop Havana, Cuba 27
Mar. 25-26 SPEEDUP Workshop Cadro-Lugano, Switz. 07
Mar. 25-27 Conference Honoring Richard Varga Kent, OH 48
Mar. 29-31 Numerical Mathematics Stellenbosch, South Africa 08

Apr. 10-11 Numerical Analysis at AMS Meeting Las Vegas, NV 46
Apr. 11-13 Computing for Financial Planning Ischia, Italy 06
Apr. 11-16 Multigrid Methods Copper Mountain, CO 08
Apr. 12-14 High Performance Numerical Methods Amsterdam, Netherlands 44
Apr. 12-15 British Applied Mathematics Colloquium Bath, UK 05
Apr. 12-16 Kinetic and Hyperbolic Problems S. Margherita Ligure, Italy 06
Apr. 15-20 Scattered Data Fitting Puerto Vallarta, Mex. 98:06
Apr. 16 Irreguarly Sturctured Problems San Juan, Puerto Rico 08
Apr. 19-21 PDE and CFD Pisa, Italy 09
Apr. 24 Midwest Numerical Analysis Day Chicago, IL 06
Apr. 29 Southern Ontario NA Day Hamilton, Ontario 08

May 7- 9 Optimization in Computational Chemistry Princeton, NJ 27
May 10-14 Theoretical and Computational Acoustics Trieste, Italy 34
May 10-12 SIAM Conference on Optimization Atlanta, GA 41
May 12-15 SIAM Annual Meeting Atlanta, GA 41
May 17-20 Conference Honoring Larry Schumaker Nashville, TN 08
May 19-22 Mathematics from Physics Urbana, IL 44
May 21-22 Differential Equations Mississippi State, MS 43
May 23-26 Parallel CFD Williamsburg, VA 07
May 24-26 Discontinuous Galerkin Methods Newport, RI 08
May 24-28 Applications of Dynamical Systems Snowbird, UT 23
May 24-31 Environmental Modeling Rostov-on-Don, Russia 07
May 31-June 4 Computer Algebra Munich, Germany 02

June 2- 6 Large Scale Scientific Computations Sozopol, Bulgaria 08
June 2- 6 Matrices with Sparsity Structure Sozopol, Bulgaria 45
June 3- 4 Mathematical Modelling and Analysis Vilnius, Lithuania 48
June 6-11 Rational Approximation Antwerp, Belgium 46
June 6-11 Rational Approximation Antwerp, Belgium 31
June 7- 9 Operational Research In Motion Windsor, Ontario 29
June 7-11 Monte Carlo Methods Varna, Bulgaria 22
June 10-12 Preconditioning Techniques Minneapolis, MN 09
June 11 Computational Methods in Engineering Uppsala, Sweden 03
June 11-13 CAIMS Annual Meeting Quebec, Canada 05
June 12-14 Java Grande Conference Palo Alto, CA 46
June 13-16 Computational Geometry Miami Beach, FL 44
June 13-19 Summer School Sozopol, Bulgaria 04
June 14-18 Householder Numerical Algebra Symposium Whistler B.C., Canada 42
June 15-20 Computational Physics Dubna, Russia 45
June 16-18 Wavelet Workshop Braunschweig, Germany 04
June 16-18 Numerical Dynamical Systems Milano, Italy 09
June 17-18 Honor of George K. Lea Arlington, VA 06
June 20-24 Number Theory Winnipeg, Canada 40
June 21-25 Matrix Theory Haifa, Israel 08
June 21-26 Continuous Optimization Rio de Janeiro, Brazil 33
June 21-27 Course on Computational Mathematics Martina Franca, Italy 05
June 21-27 Industrial Applications Apulia, Italy 35
June 22-25 Finite Elements and Applications Brunel, UK 05
June 27-July 1 Structured Matrices Boulder, CO 04
June 28-July 1 Parallel Processing and Imaging Las Vegas, NV 49
June 28-July 1 Differential Equations Las Vegas, NV 05
June 28-July 2 Numerical Analysis Dundee, Scotland 04

July 1- 4 Scientific and Enginnering Computing Beijing, China 06
July 1- 7 Curves and Surfaces Saint-Malo, France 15
July 3- 6 Accurate Solution of Eigenvalue Problems Hagen, Germany 06
July 4- 8 Circuits, Systems, Comms and Computers Athens, Greece 08
July 5- 9 ICIAM Edinburgh, Scotland 47
July 12-16 System Modelling and Optimization Cambridge, UK 44
July 15-17 Viscosity Crete, Greece 08
July 16-18 Algorithms Prague, Czech Republic 40
July 18-28 Foundations of Computational Mathematics Oxford, U.K. 44
July 19-22 International Linear Algebra Society Barcelona, Spain 36
July 19-23 Finite Volumes for Complex Applications Duisburg, Germany 06
July 26-30 Numerical Mathematics Jyvaskyla, Finland 42

Aug. 2- 6 Computational Physics Beijing, China 04
Aug. 4- 6 Semiconductor Technology Boulder, CO 02
Aug. 9-13 SciCADE99 Queensland, Australia 47
Aug. 15-18 Knowledge Discovery and Data Mining San Diego, CA 06
Aug. 16-20 Ordinary Differential Equations Auckland, New Zealand 08
Aug. 21-25 IMACS Congress 2000 Lausanne, Switzerland 46
Aug. 25-28 Conference Honoring Sergei Godunov Novosibirsk, Russia 47
Aug. 25-28 Viscoelastic Flows Vaals, Netherlands 42
Aug. 27-29 Computational Plasticity Kiel, Germany 03
Aug. 31-Sep. 3 Euro-Par'99 Toulouse, France 04

Sep. 6-10 Numerical Solution of Markov Chains Zaragoza, Spain 38
Sep. 6-10 Parallel Computing Technologies St.-Petersburg, Russia 41
Sep. 6-21 Numerical Simulation of Flows Heidelberg, Germany 09
Sep. 13-18 Applied Math and Computation Dubrovnik, Croatia 04
Sep. 13-18 Cubature Formulae Krasnoyarsk, Russia 05
Sep. 20-24 Computational Techniques Canberra, Australia 42
Sep. 20-23 Transport-Dominated Problems Magdeburg, Germany 48
Sep. 27-30 Optimization and Numerical Algebra Nanjing, China 07
Sep. 27-Oct. 1 Boundary Integral Methods Sydney, Australia 49
Sep. 28-Oct. 3 Global Optimization Firenze, Italy 05

Oct. 4- 8 Parametric Optimization Dubrovnik, Croatia 45
Oct. 4- 8 Complexity of Multivariate Problems Hong Kong 06
Oct. 9 Conference Honoring Cleve Moler Natick, MA 04
Oct. 18-22 Conference Honoring S. K. Godunov Manchester, UK 06

Nov. 14-19 Forum on Parallel Computing Methods Nashville, TN 08

Dec. 6- 8 CFD Modelling Melbourne, Australia 06
Dec. 12-17 Applied and Computational Math Chile 08


From: Ian Michael Mitchell <imitchel@leland.Stanford.EDU>
Date: Sun, 21 Feb 1999 17:01:51 -0800 (PST)
Subject: Stanford's SCCM Celebrates Decade

The Scientific Computing and Computational Mathematics (SCCM) program
at Stanford celebrated its tenth (+1) anniversary on Feb. 6, 1999 with
a one-day meeting on scientific computing. A series of talks was
organized by current SCCM director Andrew Stuart and presided over by
founding director Gene Golub (and his new laser pointer).

General talks covering various aspects of scientific computing were given by
Parvis Moin of Stanford on turbulence research,
Stan Osher of UCLA on PDE based image processing,
John Hennessy of Stanford on high performance computing and
Jim Demmel of UC Berkeley on linear algebra.

SCCM also welcomed back program graduates Liliana Borcea
(Computational Science at Rice) and Hongyuan Zha (Computer Science at
Penn State), who gave talks on their current research activities.
More than forty-five faculty, students and other researchers attended
the talks, and the day was rounded out by a dinner at the faculty club
that evening.

Ian Mitchell
SCCM Ph.D. student


From: G. W. Stewart <>
Date: Wed, 24 Feb 1999 14:37:13 -0500 (EST)
Subject: Jampack, A Package for Manipulating Matrices in Java

I am pleased to announce the initial release of Jampack, a package for
manipulating matrices in Java. The package was developed at
NIST and the University of Maryland. It has the following features.

1. Classes for general, diagonal, and triangular complex matrices.
(In the future real matrices and a band matrix class will be

2. Suites of methods to implement the common matrix operations and
functions, including the solution of linear systems.

3. Classes implementing the common matrix decompositions.

4. Classes to generate and apply plane rotations and Householder

5. A feature for keeping track of matrix decompositions so they
do not need to be recomputed when they are used later.

Jampack is not a finished package. The purpose of this preliminary
release is to find out if there is enough interest in the package
to justify its further development. Jampack and its documentation
can be obtained from the University of Maryland at

or from NIST at

G. W. (Pete) Stewart


From: Jack Dongarra <>
Date: Tue, 23 Feb 1999 15:07:09 -0500
Subject: ATLAS Version 2.0 is Now Available

ATLAS version 2.0 is now available.
To obtain a copy of the ATLAS 2.0 see:

The existing BLAS have proven to be very effective in facilitating the
production of portable, efficient software for sequential, vector and
shared memory high-performance computers. However, hand-optimized BLAS
are expensive and tedious to produce for any particular architecture,
and in general will only be created when there is a large enough market,
which is not the true for all platforms. The process of generating an
optimized set of BLAS for a new architecture or a slightly different
machine version can be a time consuming process. The programmer must
understand the architecture, how the memory hierarchy can be used to
provide data in an optimal fashion, how the functional units and
registers can be manipulated to generate the correct operands at the
correct time, and how best to use the compiler optimization.
ATLAS is an approach for the automatic generation and optimization of
numerical software for processors with deep memory hierarchies and
pipelined functional units. ATLAS has been designed to automate much of
this process. This version of ATLAS focuses on the Level 3 BLAS. ATLAS
optimizes the operations to account for many parameters such as blocking
factors, loop unrolling depths, software pipelining strategies, loop
ordering, register allocations, and instruction scheduling. ATLAS does
this by carrying out a parameter study on the target architecture and
generating the "optimal" code. ATLAS' performance wins are evident in a
number of cases. On a system where there is presently no vendor supplied
BLAS (viz. DEC ALPHA 21164 running Linux), the portable Fortran
reference implementation gets less than 8% of the performance that ATLAS
achieves. On platforms where vendors have hand-tuned this kernel
library, ATLAS produces software that runs at roughly the same rate.
ATLAS generally requires only 1-2 hours on new hardware in order to tune
itself to the same level of efficiency.
While the present release is for the Level 3 BLAS, much of the
technology and approach developed here can be generalized relatively
easily. We therefore plan to support the level 1 and 2 BLAS in the near
future. Indeed, we plan to extend the ATLAS concept to cover other
operations as well. In particular, we would like to provide sparse
linear algebra and Java language support. The release of ATLAS as well
as documentation can be found at

C. Whaley and J. Dongarra


From: Ron Shuval <>
Date: Thu, 25 Feb 1999 23:11:14 +0200
Subject: GISOLV2, A Package for Solving Linear Systems

APMATH announces the release of GISOLV2.
GISOLV2 is a package of routines based on new algorithm
for solving symmetric and non-symmetric linear systems.
It almost eliminates the stagnation phenomenon related
to other iterative algorithms and it is highly efficient
for solving sets of linear equations which share the same
matrix. A free beta version can be downloaded at


From: Sivaguru S. Sritharan <>
Date: Wed, 24 Feb 1999 13:24:00 -0800
Subject: Change of Address for S. S. Sritharan

Hello Numerical Analysis colleagues, here is my new contact information:

Dr. S. S. Sritharan, Head
Surveillance Radar Technology (Code D73H)
Joint & National Systems Division
San Diego, CA 92152-5001

Ph: 619-553-1201
Fax: 619-553-1586


From: Daniel Hershkowitz <>
Date: Wed, 24 Feb 1999 08:26:35 +0200 (IST)
Subject: New Editors for Electronic Journal of Linear Algebra

The ELECTRONIC Journal of LINEAR ALGEBRA (ELA) is happy to announce the
appointment of five new advisory editors and three new associate editors.

The new advisory editors are:

Richard A. Brualdi
Ludwig Elsner
Miroslav Fiedler
Shmuel Friedland
Hans Schneider

The new associate editors are:

Ravindra B. Bapat
Steve Kirkland
Bryan L. Shader

Our entire board is shown in ELA's primary homepage:

and in ELA's mirror sites are: (Temple University) (Univ. Lisbon) (Univ. Chemnitz) (EMIS)


From: Marcin Paprzycki <>
Date: Sun, 21 Feb 1999 00:31:43 -0600 (EST)
Subject: New Book Series, Advances in Computation

Dear Colleagues,

It is our pleasure to announce that the book series "Advances in the
Theory of Computation and Computational Mathematics" has been revived
under new name, new publisher and a considerably modified editorship. The
new series:


will be published by NOVA Science Books. Its Editorial Board consists of:

Ilan Bar-On, Technion University
Charles Dunham, University of Western Ontario
Graeme Fairweather, Colorado School of Mines
Daniele Funaro, Universita di Modena
Dan Grigoras, Technical University Iasi
Andreas Karageorghis, University of Cyprus
Michal Karonski, Adam Mickiewicz University
Erricos John Kontoghiorghes, Universite de Neuchatel
Anna Lawniczak, University of Guelph
Michael Mascagni, University of Southern Mississippi
Nikos E. Mastorakis, Hellenic Naval Academy
Lawrence F. Shampine, Southern Methodist University
Jurij Silc, Jozef Stefan Institute
Wojciech Szpankowski, Purdue University
Daniel B. Szyld, Temple University
Gyorgy Turan, University of Illinois at Chicago
Jerzy Wasniewski, Danish Computing Center, UNI-C
Plamen Yalamov, University of Rousse

We intend to publish books from any area of mathematics or computer
science involved with computation broadly understood.

We plan to publish four types of high quality manuscripts:

- monographs
- topic-oriented volumes consisting of contributed papers devoted to
the current research
- topic-oriented volumes consisting of invited overview papers,
summarizing the state of the art
- proceedings from selected Workshops/Conferences/Minisymposia

To discuss a possible contribution, please contact the Managing Editor at
the address below.


Lee Keener, Editor-in-chief
College of Science and Management
University of Northern British Columbia
Prince George, BC V2N 4Z9, CANADA

Marcin Paprzycki, Managing Editor
Department of Computer Science and Statistics
University of Southern Mississippi
Hattiesburg, MS 39406-5106, USA


From: G. Wittum <>
Date: Mon, 22 Feb 1999 00:06:50 +0100
Subject: Summer School on Numerical Simulation of Flows

EMS - WiR Summer School on
Numerical Simulation of Flows
Heidelberg, September 6-21, 1999

The numerical simulation of flows is one of the central problems in
Scientific Computing. Complexity of flow simulations is so high that
a realistic description requires sophisticated mathematical methods
and models. In particular modeling and simulation of turbulent flows,
nearly incompressible flows, and multi-phase flows are challenging
problems for mathematical models and numerical methods.
There is a strong interest in this topic by numerous groups from
mathematical modelling and numerical simulation. Recently a number
of new mathematical models and methods have been introduced which
are highly relevant for flow simulations. Amongst others these are
multiscale modelling and numerics, homogenization, finite-element
and finite volume methods, spectral and h-p discretizations, grid
adaptivity and error estimators, multigrid and conjugate-gradient type
methods and wavelets. Another field of increasing importance is the
development of methods for the visualization of flows. The numerical
simulation of flows requires cooperation of several mathematical
disciplines as Analysis, Numerics, Mathematical Physics and
Computational Science.
The European Mathematical Society (EMS) together with the research
network WiR will organize a summer school on Numerical Simulation
of Flows from Sept. 6 - 21, 1999 in Heidelberg.
The Summer school will consist from a theoretical and a practical part,
each one lasting a week. The first week (Sept. 6-10) is devoted to basic
instruction. There mathematical models and methods are presented in
lectures given by specialists. In the second week (Sept. 13-17) the
participants will work on problems posed by the lecturers in several
places. Finally the results of this work will be presented in a plenary
meeting (Sept. 20-21).

Scientific Comittee:
P. Bastian (Heidelberg), G. Dziuk (Freiburg), M. Gutknecht (Zuerich),
R. Jeltsch (Zuerich), D. Kroener (Freiburg), C.-D. Munz (Stuttgart),
R. Rannacher (Heidelberg), W. Rodi (Karlsruhe), G. Wittum (Heidelberg),
H. Yserentant (Tuebingen)

Local Organizer:
Prof. Dr. G. Wittum, IWR, Technische Simulation, Universitaet
Heidelberg, Im Neuenheimer Feld 368, D-69120 Heidelberg,


From: Pavani Raffaella <>
Date: Tue, 23 Feb 1999 12:58:14 +0100
Subject: Workshop in Milan on Numerical Dynamical Systems

June 16-18, 1999

Luca Dieci (Georgia Institute of Technology)
Peter Kloeden (Goethe University - Frankfurt am Main)
Arieh Iserles (Cambridge University)

Laura Gotusso (Politecnico di Milano)
Raffaella Pavani (Politecnico di Milano)
Sandro Salsa (Politecnico di Milano)

The topics include:
- Smooth orthonormal factorization of matrices with applications to
computation of etheroclinic orbits and Lyapunov exponents
- Nonautonomous attractors and their numerical discretization
- Numerical analysis of differential equations in Lie groups

More details can be found on the WEB page

Further information and second announcement at


From: Fabio Schoen <>
Date: Tue, 23 Feb 1999 15:34:05 +0100
Subject: Workshop in Florence on Global Optimization

(GOdot 99)
International Workshop on Global Optimization
Firenze (Italy), September 28th - October 3rd 1999

The International Workshop on Global Optimization which will take place
in Florence, Italy. But we have to remind you that the deadline for
pre-registration is February 28. Please, if you plan to attend (we hope so!),
check the home page
or its mirror at


From: Francois Beux <>
Date: Thu, 25 Feb 1999 19:26:04 +0100
Subject: Workshop in Pisa on PDE and CFD

Workshop on:
Advances in Stabilized Methods for Partial Differential
Equations with Emphasis on Fluid Dynamics
April 19 -21, 1999 Scuola Normale Superiore Pisa, Italy

This workshop is sponsored by the "Associazione Amici della Scuola
Normale Superiore di Pisa" and by ESF through the program AMIF

Scientific direction:

Prof. G. Da Prato - Scuola Normale Superiore
Prof. T.J.R. Hughes - Stanford University
Prof. A. Quarteroni - Politecnico Milano/EPFL Lausanne

Organized by

T. J. R. Hughes

The Mary and Gordon Crary Professor of Engineering
Chair, Mechanics and Computation Stanford University
Stanford, California 94305 U. S. A.

The workshop will consist of invited lectures given by nine
leading international expert on recent advances in the theory and
application of stabilized methods.

The following topics will be covered :
C. Canuto, Politecnico of Torino
- Stabilization of non-coercive problems by multilevel devices
L. Franca, Univ. of Colorado - Denver
- What are residual free bubbles?
- Beyond residual free bubbles: Two-level finite element
methods in computational fluid dynamics (CFD)
T. Hughes, Stanford University
- The variational multiscale method in CFD
C. Johnson, Chalmers Institute - Goteborg
- Adaptive error control in CFD
E. Onate, Polit. Univ. Of Cataluna - Barcellona
- Stabilization of numerical solution of advective-diffusive and
fluid flow problems via a finite increment calculus approach
A. Quarteroni, Politecnico Milano /EPFL Lausanne
- Algebraic splitting methods for Navier-Stokes equations
Multifield methods for flow problems
A. Russo, Istituto An. Numerica, Pavia
- Stabilization of hyperbolic problems via residual-free bubbles
E. Suli, Oxford University
- A posteriori error analysis of stabilised finite element methods
for hyperbolic problems
T. Tezduyar, Houston University
- Computational methods for three-dimensional simulation of
fluid-object interaction
- Computational methods for three-dimensional simulation of
free-surface flows

FOR FURTHER INFORMATION (AMIF grant, registration form, ...)

please see the web page:

or contact us

Associazione Amici della Scuola Normale Superiore
Piazza dei Cavalieri, 7 - 56126 Pisa
Tel.(+39) 050 509058 Fax (+39) 050 563513
pag. web:


From: Yousef Saad <>
Date: Tue, 23 Feb 1999 11:43:23 -0600 (CST)
Subject: Deadline Extended for Sparse 99


June 10 - 12, 1999

University of Minnesota, Hubert H. Humphrey Institute,
Minneapolis, Minnesota


Please see previous announcements in na-digest or consult the following
web-site for details:


From: Christoph Ueberhuber <>
Date: Wed, 24 Feb 1999 11:20:06 +0100
Subject: Postdoctoral Position at Technical University of Vienna

The Institute for Applied and Numerical Mathematics of the
Technical University of Vienna (Austria) has a post-doctoral
research fellowship available starting on or around April 9, 1999.

The institute has ongoing research activities in a number of areas
of computational mathematics. Our interests range from applied analysis
to the development of numerical algorithms and software for a variety
of advanced computer architectures.

DUTIES: The appointee will participate in a research effort implementing
large scale matrix problems on single-processor and multi-processor com-
puter systems. The position requires significant programming and technical
interaction with other members of the research team and the scientific
computing community in Vienna.
The successful applicant must be able to work with and extend existing
numerical software and develop reliable and readable code. Throughout
the project, the appointee should be able to investigate issues of
algorithmic design. She or he will also be expected to prepare results for
publication in scientific journals and for presentations at scientific

QUALIFICATIONS: Essential --Ph.D. in applied mathematics or computer
science. Experience with the development of linear algebra library
software on parallel architectures is highly desirable. Experience
with Fortran, BLAS, LAPACK, ScaLAPACK would be very helpful.
Ability to summarize research results for publication and presentations.
Ability to interact and collaborate with other members of the research
team and other research groups.

It is a full-time, one-year appointment with the possibility of renewal.
Salary is on University's Research Associate level.

Qualified individuals are invited to submit application materials
(c.v., publication list, and names of three references) by e-mail
or Fax to:

Christoph W. Ueberhuber

Institute for Applied and Numerical Mathematics
Technical University, Vienna

Wiedner Hauptstrasse 8-10 / 115-2
A-1040 Vienna

Tel.: +43/1/ 588 01 11512
Fax: +43/1/ 588 01 11599

Applications should be submitted as soon as possible and will be
accepted until the position is filled.


From: Jun Zhang <>
Date: Wed, 24 Feb 1999 11:44:28 -0500 (EST)
Subject: Graduate Research Assistantships at University of Kentucky

Two Graduate Research Assistantships in
High Performance Scientific Computing

Two graduate research assistantships in high performance
scientific computing and computational mathematics are
available in the Department of Computer Science at the
University of Kentucky for highly motivated Ph.D.
students starting Fall 1999.

The candidates need to be admitted into the Graduate
Program in the Computer Science Department and will join
Jun Zhang's high performance scientific computing research
group consisting of a number of postdoctoral and graduate
students. They should have solid background in applied
numerical analysis and computational (programming) skills.
Research topics include, but are not limited to:

1.) high performance robust preconditioning techniques for
general sparse linear systems; and
2.) robust parallel multigrid methods for CFD applications.

For detailed information, please check the web page at

Interested person should e-mail a copy of his/her curriculum
vitae in postscript or ASCII to Jun Zhang at Otherwise please fax it to (606)323-1971
or send a postal mail to:

Professor Jun Zhang
Department of Computer Science
University of Kentucky
773 Anderson Hall
Lexington, KY 40506-0046


From: Chaman Singh Verma <>
Date: Thu, 25 Feb 1999 12:56:41 +0530 (IST)
Subject: Positions at Center for Development of Advanced Computing, India

Scientific and Engineering Computing Positions
Center for Development of Advanced Computing, Pune, INDIA

Scientific and Engineering computing group at C-DAC is seeking highly
motivated people for the following projects.

1. Development of Parallel Incompressible Flow simulation.
2. Development of Parallel Visualization algorithms for
Computational Fluid Dynamics and Seismic data.
3. Development of Parallel Seismic data processing algorithms.
4. Development of Parallel Solvers for Partial Differential Equations.

Experience required:

Visualization Projects : C/C++, MPI, OpenGL and Spatial Data structures.
Flow Simulation : F77/F90 ,Fluid Mechanics and iterative solvers.
Seismic Data Processing: F77/C, MPI and Geophysics.
PDE : Strong backgroud in Mathematics and expertise in
integration of different modules for commercial

Eligibility is limited to applicants with B.Tech( with high academic records),
M.Tech/Ph.D in Computer Science, AeroSpace Engineering, Mechanical, Geophysics
or related areas.

Because of Government of India rules, we are seeking only Indian citizens.
The projects are likely to start from May 1, 1999.

Please send a copy of your resumes latest by 20 March, 1999.

Dr. Dheeraj Bhardwaj/ Chaman Singh Verma,
Center for Development of Advanced Computing (C-DAC),
Pune University Campus,Ganeshkhind
Pune-411007 ( Maharastra) INDIA
E-mail :
E-mail :


From: Scott Baden <>
Date: Thu, 25 Feb 1999 09:36:53 -0800 (PST)
Subject: Research Position at U. C. San Diego

Scientific Computation Group
Department of Computer Science and Engineering
University of California, San Diego
Scott B. Baden,

The KeLP Project in the Department of Computer Science and Engineering at
UC San Diego ( is looking for
a staff member to play a lead role in developing and evaluating run time
software techniques for compute- and data- intensive scientific
applications on high performance parallel computers. The position is
funded by the NPACI.

The duties for this position include: point-of-contact for KeLP users for
supported applications; evaluating and implementing performance
enhancements and extensions; retargeting KeLP to emerging computer
architectures such as the NPACI IBM SP Tflops machine due to arrive later
this year. Prior parallel computing experience is required.

For a full description of the requirements see the URL and look for position C3504-S

This position is part of the SOURCE...UCSD Hiring System Pilot and requires
a resume and resume supplement. Please do not fax resumes for this
position. If you have previously applied to become part of the SOURCE
pilot, you will automatically be considered for this position and do not
need to reapply. See for more
information. UCSD is an equal opportunity employer.


From: P. K. Jimack <>
Date: Thu, 25 Feb 1999 17:59:22 GMT
Subject: Lectureship Positions at University of Leeds

School of Computer Studies, University of Leeds

Two permanent lectureships are currently available in the School
and applications are invited from candidates with any of a wide
range of reseaarch interests, including Scientific Computation.

Preference will be given to candidates who strengthen our existing
research activities:

* Artificial Intelligence (automated reasoning, computer vision,
natural language processing)

* Information Systems (information modelling and management)

* Internet Computing (virtual working systems, virtual prototyping,
networking, scalable high-performance systems, visualization)

* Scheduling and Constraint Management (transport scheduling, linear
programming, metaheuristics, constraint programming)

* Scientific Computation (computational PDEs, parallel algorithms)

* Theoretical Computer Science (algorithms and complexity, formal
methods, safety-related systems)

You should have a PhD (or equivalent experience) in a relevant
discipline. Prior experience in teaching, preferably in higher
education, would be an advantage.

Salary: Lecturer A/B scales (currently 16,555 to 29,048 ukp p.a.)
according to qualifications and experience.

Further details may be obtained from the School's Web Pages at:


From: J. Sun <>
Date: Sat, 27 Feb 1999 11:55:53 +0800
Subject: Graduate Fellowships at Singapore-MIT Alliance

Applications are invited for admission to the following full-time Graduate
courses at National University of Singapore (NUS) in the 1999-2000 academic
year (commencing 1 July 1999):

1. Master of Science in Advanced Materials
2. Master of Science in High Performance Computation for Engineered Systems


3. Master of Engineering
4. Doctor of Philosophy

Programmes 1 & 2 are based on course work while Programmes 3 & 4 are based
on research and course work as well as a thesis.


All successful applicants for any one of the above programmes will be
awarded a scholarship which will carry monthly emoluments of S$1,400.
Top-up grants will be available for the Master of Engineering and Doctor of
Philosophy programmes. In addition, the tuition fee for all the programmes
will be waived.


The period of candidature for Programmes 1 & 2 is one year, which comprises
two semesters. ( One semester is 17-18 weeks ). All candidates will be
required to attend two weeks of intensive classes ( Immersion Programme)
at the Massachusetts Institute of Technology ( MIT) in the U.S.

The minimum period of candidature for Programme 3 is one and a half years.
The minimum period of candidature for Programme 4 is three years.
All candidates for Programmes 3 & 4 will be required to spend a term at the
MIT campus.


Selected NUS/Nanyang Technological University (NTU) and MIT Professors will
jointly conduct courses offered by SMA. MIT Professors will be at the NUS
campus for a certain period of time throughout the duration of the
programmes and continue to interact with students via long distance learning
when at MIT via state-of-the-art IT and communications technology.


Application forms and other information can be obtained personally from the
SMA Office at Engineering Block 4, 6th Floor, Room 21 or in writing to :

The Director
Singapore-MIT Alliance (SMA)
National University of Singapore
10 Kent Ridge Crescent
Singapore 119260

Fax no. : (65) 775 2920
E-mail address :
Homepage address :


The closing date for receipt of applications is 25 March 1999.


From: Thomas Hogan <>
Date: Tue, 23 Feb 1999 15:27:01 -0500 (EST)
Subject: Contents, Journal of Approximation Theory

Table of Contents: J. Approx. Theory, Volume 96, Number 2, February 1999

Ernst G\"orlich and Alfred P. Rohs
Polynomial projections in $C[-1,1]$ and $L^1(-1,1)$ with growth
$n^\gamma$, $0<\gamma\le 1/2$

Sigrid Fredenhagen, Hans Joachim Oberle, and Gerhard Opfer
On the construction of optimal monotone cubic spline interpolations

D. Legg and Y. Pan
Discontinuity of best harmonic approximants

Y. K. Hu, K. A. Kopotun, and X. M. Yu
Weak copositive and intertwining approximation

Vladimir N. Sorokin and Jeannette Van Iseghem
Matrix continued fractions

Bl. Sendov
Adaptive multiresolution analysis on the dyadic topological group

X.-S. Jin and R. Wong
Asymptotic formulas for the zeros of the Meixner polynomials

Renato Spigler, Marco Vianello, and Francesca Locatelli
Liouville-Green-Olver approximations for complex difference equations

Luis Bernal-Gonz\'alez
Hypercyclic sequences of differential and antidifferential operators

Simon J. Smith
On the positivity of the fundamental polynomials for generalized
Hermite-Fej\'er interpolation on the Chebyshev nodes

H. T. Koelink
Some basic Lommel polynomials

L. Bos and S. De Marchi
Limiting values under scaling of the Lebesgue function for polynomial
interpolation on spheres

Francisco Luquin and Concepci\'on Besga
Estimation of norms of multivariate polynomials with integral

D. S. Lubinsky
On boundedness of Lagrange interpolation in $L_p$, $p<1$


Dawei Shen
On the coefficients and zeros of a polynomial

Author index for Volume 96


From: Ake Bjorck <>
Date: Fri, 26 Feb 1999 16:56:48 +0100 (MET)
Subject: Contents, BIT

CONTENTS BIT Volume 39, No. 2 (June 1999) ISSN 0006-3835

Optimal multipliers for linear congruential pseudo random
number generators with prime moduli: Parallel
computation and properties
D. BRUNNER and A. UHL 193

Backward error bounds for constrained least squares
A. J. COX and N. J. HIGHAM 210

Quasi-Newton methods for nonlinear least squares focusing
on curvatures

How to solve nonlinear equations when a third order method
is not applicable

Stability analysis of Runge--Kutta methods for non-linear
delay differential equations
C. HUANG, H. FU, S. LI, and G. CHEN 270

On the implementation of the method of Magnus series for
linear differential equations

The sensitivity of a spline function to perturbations of
the knots
T. LYCHE and K. M\O RKEN 305

Block-projections algorithms with blocks containing
mutually orthogonal rows and columns
C. POPA 323

The Bezout number for piecewise algebraic curves
X. SHI and R. WANG 339

Remarks on non-linear spectral perturbation


Bounds for the positive root of a class of polynomials
with applications

A note on an error bound for the AOR method
Y. SONG 373


End of NA Digest