- Today's Editor:
- Cleve Moler
- The MathWorks, Inc.
- moler@mathworks.com

- Solving A*X=B When Only B is Known
- Change of Address for Ed Saff
- NetSolve Version 1.4 Available
- Kestrel Interface to NEOS Server
- Symposium in Nanoi on Generalized Convexity/Monotonicity
- Workshop at LBNL on Advanced Computational Testing and Simulation
- Workshop in Barcelona on Agro-food Technology
- Conference in Perth on Boundary and Interior Layers
- Conferences in Boston on Imaging Science and Life Sciences
- Workshop in Knoxville on Linear Systems Solving in CFD
- Short Course on Error Estimation and Adaptive Discretization
- Conference in Mexico on Domain Decomposition Methods
- Faculty Position at University of Leicester
- Postdoctoral Position at Australian Nuclear Science Organisation
- Contents, Reliable Computing
- Contents, Numerical Linear Algebra with Applications
- Contents, Journal of Computational Analysis and Applications

**URL for the World Wide Web:**
http://www.netlib.org/na-net/na_home.html

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

From: Jeff Canfield <canfield@rbc6000.scs.uiuc.edu>

Date: Fri, 3 Aug 2001 17:52:23 -0500 (CDT)

**Subject: Solving A*X=B When Only B is Known**

I am trying to solve the linear algebra problem A*X=B with the following

constraints:

1) A is an unknown s-by-c real matrix with all elements >=0

2) X is an unknown c-by-n real matrix with all elements >=0

3) B is a known s-by-n real matrix with all elements >=0

4) each row of A sums to 1; that is, A times a column vector of 1's

gives a column vector of 1's

Some additional constraints that may help ensure a unique solution are the

following:

5) the 1st row of A is all 0's except for a 1 in column 1

6) the 1st row of X and the 1st row of B are all 1's

7) the right-most columns of X and B are all 1's

So far my approach has been to use an SVD (singular value decomposition)

of B (B=U*D*V' where V' is the transpose of V) to get an A and X matrix

that solve A*X=B and then to use another matrix Z (there are infinitely

many to choose from) and its inverse (inv(Z)) to convert A and X into a

form that still solves A*X=B but that also satisfies various constraints.

Basically, A=U*Z while X=inv(Z)*D*V'.

If you can suggest better ways to approach this problem (ones that will

reliably give a unique solution for A and X) or suggest reliable

algorithms that already exist to solve such a problem, please let me know.

Thank you very much,

Jeff Canfield

Department of Physics

Emory University

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

From: E. B. Saff <esaff@math.usf.edu>

Date: Tue, 31 Jul 2001 16:27:51 -0400 (EDT)

**Subject: Change of Address for Ed Saff**

Change of coordinates for Ed Saff:

E.B. Saff

Department of Mathematics

Vanderbilt University

Nashville, TN 37240

USA

email: esaff@math.vanderbilt.edu

phone: (615) 322-2014

fax: (615) 343-0215

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

From: Jack Dongarra <dongarra@cs.utk.edu>

Date: Fri, 03 Aug 2001 13:52:33 -0400

**Subject: NetSolve Version 1.4 Available**

NetSolve grid based software system, version 1.4 is now available.

The software can be downloaded from: http://icl.cs.utk.edu/netsolve

The NetSolve project is being developed at the University of Tennessee's

Innovative Computing Laboratory. The system provides NetSolve-enabled programs

with transparent network access to computational resources via its many

client application programming interfaces or APIs. In this version, client

programs implemented in C, Fortran, Matlab and Mathematica can access the

NetSolve system and the hardware and software services it provides. These

services include sophisticated numerical solvers from libraries like LAPACK,

ScaLAPACK, PETSc, Aztec, SuperLU, etc. Facilities are available for creating

new NetSolve services. All components have been tested on a variety of

UNIX operating systems, including Linux, AIX, Irix, OSF, and Solaris. A

client interface is available for the Microsoft Windows 2000 platform.

There is a test grid deployed worldwide for experimentation that is accessible

by anyone who has installed the new client interface.

See http://icl.cs.utk.edu/netsolve/ for further details and documentation.

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

From: Elizabeth Dolan <dolan@mcs.anl.gov>

Date: Mon, 30 Jul 2001 17:36:00 -0500

**Subject: Kestrel Interface to NEOS Server**

The Optimization Technology Center is pleased to announce the public

release of the Kestrel interface to the NEOS Server. Individuals

with local access to either the AMPL or GAMS modeling languages can

use the Kestrel client to solve models remotely through the NEOS

Server in much the same way they would solve a model on a local

machine. Results from the remote solve are returned to the modeling

language in its native format for further manipulation. More

information on Kestrel and the client executables for various platforms

are available from http://www-neos.mcs.anl.gov/neos/kestrel.html

Questions or comments about the Kestrel interface to the NEOS Server

should be directed to <neos-comments@mcs.anl.gov>.

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

From: Juan Enrique Martinez Legaz <jemartinez@selene.uab.es>

Date: Tue, 31 Jul 2001 19:12:09 +1000

**Subject: Symposium in Nanoi on Generalized Convexity/Monotonicity**

FIRST ANNOUNCEMENT

7th International Symposium on Generalized Convexity/Monotonicity

The Hanoi Institute of Mathematics has the honor of hosting the 7th

International Symposium on Generalized Convexity/Monotonicity which will

be held on August 27-31, 2002 in Hanoi/Vietnam.

The conference is organized by the international Working Group on

Generalized Convexity (WGGC) (http://genconv.ec.unipi.it) and sponsored

by the Pacific Optimization Research Activity Group (POP)

(http://www.polyu.edu.hk/~ama/links/pop/an1.html). Previous conferences

in this research area took place in Vancouver-Canada 1980, Canton-USA

1986, Pisa-Italy 1988, Pecs-Hungary 1992, Luminy-France 1996 and

Samos-Greece 1999. The symposium is aimed at bringing together

researchers from all continents to report their latest results and to

exchange new ideas in the field of generalized convexity and generalized

monotonicity and its applications in optimization, control,

stochastics, economics, management science, finance, engineering and

related topics.

The scientific program will consist of four invited lectures of 45

minutes (the full list of invited speakers will shortly be announced)

and contributed talks of 30 minutes.

Participants from all countries are welcome to attend and are encouraged

to present contributed talks.

Program Committee :

Martinez-Legaz, J.-E. (Barcelona, Spain)(co-chairman)

Sach, P.H. (Hanoi, Vietnam)(co-chairman),

Cambini, R. (Pisa, Italy)

Crouzeix, J.-P. (Clermont-Ferrand, France)

Eberhard, A. (Melbourne, Australia)

Hadjisavvas, N. (Samos, Greece)

Komlosi, S. (Pecs, Hungary)

Luc, D.T. (Avignon, France, and Hanoi, Vietnam)

Schaible, S. (Riverside, USA)

Addresses to Contact :

NGUYEN DINH CONG DINH THE LUC

Institute of Mathematics

P.O.Box 631, Boho

10000 HANOI, Vietnam

Tel. (84) 4 7 563 474

Fax. (84) 4 7 564 303

E-mail : gcm7@thevinh.ncst.ac.vn

Departement de Mathematiques

Universite d'Avignon, 33 rue Louis Pasteur

84000 AVIGNON, France

Tel. (33) 4 90 86 36 59 ; (33) 4 90 14 44 10

Fax. (33) 4 90 14 44 19

E-mail : dtluc@univ-avignon.fr

A more detailed announcement containing information on deadlines and

other matters will be available on line at http://genconv.ec.unipi.it

and http ://203.162.7.82/

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

From: Tony Drummond <LADrummond@lbl.gov>

Date: Tue, 31 Jul 2001 13:53:27 -0700

**Subject: Workshop at LBNL on Advanced Computational Testing and Simulation**

NERSC INVITES COMPUTATIONAL SCIENTISTS TO ATTEND OCT. 10-13 WORKSHOP

IN BERKELEY ON ADVANCED COMPUTATIONAL TESTING AND SIMULATION

BERKELEY, Calif. - The U.S. Department of Energy's National Energy Research

Scientific Computing Center (NERSC) is hosting a three and half-day workshop

on Advanced Computational Testing and Simulation (ACTS) on Oct. 10-13 at

Lawrence Berkeley National Laboratory.

The ACTS workshop, "Tools for Advanced Computational Testing and Simulation

- Solving Problems in Science and Engineering," is aimed at familiarizing

researchers in various scientific disciplines with the computational science

tools now available in the DOE's ACTS Toolkit. Therefore, the workshop will

include a range of tutorials on the tools, discussion sessions focused on

solving specific computational needs by the participants, and hands-on

practices using the NERSC's state-of-the-art computers.

The ACTS tools have already provided developers and users of scientific

and engineering applications with solutions that help their research. The

services provided by these tools range from implementations of numerical

algorithms, scientific data representations, data manipulation, remote

visualization, program execution and distributed computing. For more

information visit the ACTS information center on the Web at

http://acts.nersc.gov.

There are no registration fees for the workshop, but those interested in

attending must complete the on-line application form at

http://acts.nersc.gov/workshop/application.html and include an outline of

their current work and future plans and needs for computational resources.

The deadline for filing applications is Friday, Aug. 31.

In addition, DOE will provide financial support to enable a limited

number of graduate students and post-docs to attend this workshop. Graduate

students and post-docs from any scientific field who are interested in

attending the workshop must complete the on-line application form and submit

a short abstract describing their work and current or future computational

needs. Each post-doc or graduate student applicant must also provide a letter

of reference, which can be submitted via the Web at

<http://acts.nersc.gov/workshop/recommendation.html/> by his or her advisor.

The recommendation letter must also arrive no later than August 31, 2001.

For more information on the workshop, go to

<http://acts.nersc.gov/workshop/> or contact Tony Drummond at (510) 486-7624

or Osni Marques at (510) 486-5290.

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

From: Julio Banga <julio@iim.csic.es>

Date: Wed, 01 Aug 2001 11:57:36 +0200

**Subject: Workshop in Barcelona on Agro-food Technology**

First International Workshop on

"MATHEMATICAL AND COMPUTING TECHNIQUES

FOR AGRO-FOOD TECHNOLOGIES"

26-27 November 2001

Barcelona, Spain

http://www.cimne.upc.es/congress/food/

Agro-food Technology (AfoT) is a thematic area within MACSI-Net, an

European Network supported by the Information Society Technologies

Programme (IST) of the Fifth Framework Programme of the European

Commission. MACSI-Net is an initiative to form an open network for the

advancement of Mathematics, Computing and Simulation for industry.

OBJECTIVES

The main aim of AfoT Workshop is to provide an introduction into

the most important issues of the mathematical and computing

techniques involving food technologies. These areas should contain

all the conventional mathematical modelling and computer simulation

techniques as well as signal processing methods.

The workshop is one of the activities of the MACSI-Net network to

impulse unified mathematical and computing techniques involving

food scientists, engineers and industrial people, as well as to

encourage new cooperations at international level between companies

and research institutions.

WORKSHOP TOPICS

The programme will consist of keynote and invited lectures delivered by

relevant scientist on the field. Programme will cover the following topics:

- Modelling and Simulation of Unit Operations and Process Plants

- Food Process Optimisation, Scheduling and Control

- Food Properties Measurements and Quality Control

- Simulation of Complex Processes (e.g. those requiring computational

fluid dynamics, CFD)

PRELIMINARY LIST OF LECTURES (alphabetical order of speakers)

Antonio A. Alonso "Accelerating the simulation of complex food

processes", Universidad de Vigo, Spain

Julio R. Banga "Global optimisation in food process engineering",

IIM-CSIC, Spain

Ashim K. Datta "Integration of tools for computer-aided food process

engineering", Cornell University, USA

C=E9sar de Prada "Advanced control and supervision in the sugar

industry", Universidad de Valladolid, Spain

Peter J. Fryer "Seeing flows in processes: combination of positron

emmitting particle tracking and modelling of flows inside food plant",

University of Birmingham, UK

Laurie Hall "MRI: Non-invasive, 3D quantitation of food properties/texture

and of heat/mass transport", University of Cambridge, UK

Mike McCarthy "Integration of tomographic data in process analysis,

modeling and control", University of California, USA

Bart Nicola=EF, "Simulating uncertainty in food processing",

Katholieke Universiteit Leuven, Belgium

Fernanda Oliveira "Application of mathematical modelling and computer

simulation to the design of modified atmosphere packages accounting for

product variability", University College Cork, Ireland

R. Paul Singh "Simulating food processes involving moving boundary=

problems", University of Califormia, USA

More information at:

http://www.cimne.upc.es/congress/food/

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

From: Song Wang <swang@maths.uwa.edu.au>

Date: Thu, 02 Aug 2001 10:04:53 +0800

**Subject: Conference in Perth on Boundary and Interior Layers**

BAIL 2002

An International Conference on Boundary and Interior Layers -

Computational and Asymptotic Methods

Dates: 8th to 12 July, 2002.

Location: The University of Western Australia, Perth, Australia.

Topics:

Any aspects in computational and analytical methods for boundary and interior

layers, and singular perturbation problems.

Brief history:

BAIL is a series of international conferences on Boundary and Interior Layers -

Computational and Asymptotic Methods. Seven BAIL conferences were held around

the world from 1980 to 1994. It has been decided recently by a group of

colleagues to restart the series. In accordance with this decision BAIL 2002

will be held in Perth, Australia.

International Steering Committee:

I Boglaev, Palmerston North

J Cousteix, Toulouse

P M Gresho, Livermore

B Y Guo, Shanghai

P W Hemker, Amsterdam

R B Kellogg, Columbia

J Mauss, Toulouse

J J H Miller, Dublin (Chairman)

R E O'Malley Jr, Seattle

O Pironneau, Paris

H G Roos, Dresden

R D Russell, Vancouver

Z C Shi, Beijing

G I Shishkin, Ekaterinburg

S Wang, Perth

Local Organizing Committee:

N. Fowkes, Univ of Western Australia

D. Hill, Univ of Western Australia

L.S. Jennings, Univ of Western Australia

G. Keady, Univ of Western Australia

A. Mees, Univ of Western Australia

P.-F. Siew, Curtin University of Technology

S. Wang, Univ of Western Australia (Chairman)

Y.H. Wu, Curtin Univ. of Technology

Further detailed information will be available at

http://www.maths.uwa.edu.au/~swang/BAIL2002

email:swang@maths.uwa.edu.au

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

From: Darrell Ross <ross@siam.org>

Date: Thu, 02 Aug 2001 13:37:48 -0400

**Subject: Conferences in Boston on Imaging Science and Life Sciences**

The SIAM Conferences on Imaging Science (IS01) and Conference on the

Life Sciences (LS01) are still taking preregistrations at:

http://www.siam.org/meetings/is01/

and

http://www.siam.org/meetings/ls01/

Deadline for Preregistration is August 10, 2001

The Imaging Science conference will be held from September 22-24, 2001

and the Life Sciences conference will be held from September 24-26,

2001. Both conferences will take place at the Boston Park Plaza

Hotel, Boston, MA.

For other SIAM Conferences, please visit:

http://www.siam.org/meetings/calender.htm

Darrell Ross

SIAM, Conference Program Manager

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

From: Victor Eijkhout <eijkhout@cs.utk.edu>

Date: Thu, 02 Aug 2001 16:53:35 -0400

**Subject: Workshop in Knoxville on Linear Systems Solving in CFD**

Workshop on Linear Systems Solving in CFD

Knoxville TN, 20-21 September 2001

The Innovative Computing Laboratory of the University of Tennessee at

Knoxville, in conjunction with Mississippi State University and the

Aeronautical Systems Center Major Shared Resource Center, is

organizing a two-day workshop on the solution of linear systems

arising in computational fluid dynamics.

The objective of the workshop is to provide a platform from which

strategic DoD, industrial, and academic researchers in CFD and linear

systems solution techniques may communicate their needs, ideas, and

achievements. The specific goals of the workshop are to provide an

overview of state-of-the-art algorithms and libraries for solving

large linear systems of equations arising in CFD applications, and an

overview of methods and software for matrix reordering and mesh

partitioning.

Target audience is program/project managers, code developers, and researchers.

Topics covered will include:

Iterative methods and preconditioners

Direct solvers

Mesh generation

Multigrid methods and Algebraic Multigrid

Parallelism

Invited speakers include:

Howard Elman (University of Maryland)

Anshul Gupta (IBM)

David Keyes (Old Dominion University)

Youcef Saad (University of Minnesota)

John Ruge (Front Range Associates)

The workshop committee is:

Victor Eijkhout (ICL)

Shirley Moore (ICL)

Scott Wells (ICL)

Richard Luczak (UTK - ASC MSRC)

Hugh Thornburg (MSU - ASC MSRC)

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

From: Timothy Barth <barth@nas.nasa.gov>

Date: Fri, 3 Aug 2001 09:15:45 -0700 (PDT)

**Subject: Short Course on Error Estimation and Adaptive Discretization**

NATO-RTO/NASA/VKI Course Final Announcement

ERROR ESTIMATION AND SOLUTION ADAPTIVE DISCRETIZATION

As computational fluid dynamics (CFD) is applied to ever more

demanding fluid flow problems, the tasks of (1) computing numerical

fluid flow solutions to a user specified tolerance and (2) quantitative

assessment of existing numerical fluid flow solutions have become

paramount in the design of complex fluid dynamical systems.

The goal of the NATO Research and Technology Office (RTO) sponsored

lecture series is to provide a series of comprehensive lectures by

leading experts discussing recent advances and technical progress in

the area of numerical error estimation and adaptive discretization

methods with specific emphasis on computational fluid dynamics. The

lectures are intended to accommodate attendees of both novice and

advanced levels of technical expertise.

The week long lecture series will be given at NASA Ames in the

United States and repeated later at the von Karman Institute in

Belgium. Online registration forms and hotel/travel information

is available at the WEB locations given below. Detailed lecture

notes will be available to attendees at the time of the lecture

series.

Course Lecturers and Topics

- Marshall Bern (Xerox PARC, USA): Delaunay triangulation, subdivision

surfaces, computational geometry, optimal triangulations,

adaptive refinement, mesh improvement

- Mike Giles and Endre Suli (Oxford University, UK): Introduction to

a posteriori error estimation, Giles/Pierce theory, stabilized FEM

for hyperbolic problems, a posteriori error analysis for $hp$ FEM

- Claes Johnson and Johan Hoffman (Chalmers University, Sweden): Adaptive

FEM for fluid flow, model adaptivity, multi-adaptive space-time solvers

- Jaime Peraire and Anthony Patera (MIT, USA): Implicit A-posteriori

computation of bounds, "Energy" norms and outputs of interest,

constrained minimization formulations, computation of bounds using

inexpensive relaxations

- Serge Prudhomme (University of Texas at Austin, USA): "Goal oriented"

error estimation and adaptation, a posteriori error estimation,

$hp$ FEM, stability and error control, solution adaptivity

First location

September 10-14, 2001

NASA Ames Research Center

Moffett Field, California, USA

Online Registration Available:

http://www.nas.nasa.gov/~barth/lecture_series.html

Registration DEADLINE: Aug. 31, 2001

Second location

October 15-19, 2001

von Karman Institute for Fluid Dynamics

Rhode-Saint-Genese, Belgium

Online Registration Available:

http://www.vki.ac.be (click "lecture series")

Registration DEADLINE: Oct. 1, 2001

Lecture Series Directors and Technical Contacts

Dr. T. Barth Prof. H. Deconinck

NASA Ames Research Center von Karman Institute

M.S. N202A-1 72, Chaussee de Waterloo

Moffett Field, CA, 94040 USA 1640 Rhode-Saint-Genese, Belgium

(barth@nas.nasa.gov) (deconinck@vki.ac.be)

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

From: David Keyes <keyes@icase.edu>

Date: Fri, 3 Aug 2001 13:15:43 -0400 (EDT)

**Subject: Conference in Mexico on Domain Decomposition Methods**

CALL FOR PAPERS -- deadline 31 August 2001

14th International Conference on Domain Decomposition Methods

Cocoyoc, Mexico

6-11 January 2002

http://132.248.182.141/dd14/

Domain decomposition ("divide-and-conquer") has served as an organizing

principle for many concepts and methodologies in mathematics, computer

science, and engineering. The objective of this conference is to promote

understanding and use of domain decomposition methods for the solution of

problems arising in various fields of science and engineering and to

promote interaction between researchers throughout these disciplines.

The conference will include invited plenary talks by leading experts in

the field from academia, research institutions, and industry, as well as

minisymposia, contributed papers, and graduate student paper sessions.

Conference Themes:

Analysis of Domain Decomposition Methods

Domain Decomposition Methods in Science and Engineering

Multilevel and Multiresolution Methods

Advanced Discretizations and Domain Decomposition

Mathematical Modeling and Domain Decomposition

Parallel Implementation of Domain Decomposition Methods

Demonstrations and Evaluations of Large-Scale Codes

Visit the URL above for additional information on welcome topics for

contributed papers, conference organization, registration, and logistics.

Visit http://www.ddm.org for the history and archives of conference

series.

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

From: J. Levesley <jl1@mcs.le.ac.uk>

Date: Tue, 31 Jul 2001 13:01:35 +0100 (BST)

**Subject: Faculty Position at University of Leicester**

Department of Mathematics & Computer Science

University of Leicester

England

Lecturer A in Mathematics

Available for 1 year from 1 September 2001

=A320,066 to =A323,954 pa

Ref: A5449

Applicants for this fixed term lectureship should have teaching experience

and be engaged in a strong research programme in some areas of Mathematics.

Preference may be given to applicants whose research interests complement

those of existing members of the Department. The successful applicant will

contribute to the teaching of the Department, which may include the

lecturing of a final year module related to differential geometry. Further

details about the Department may be found at http://www.mcs.le.ac.uk.

Applications close on 13 August 2001

Application forms and further particulars are available from the Personnel

Office, tel 0116 252 5114, fax 0116 252 5140, jobs@le.ac.uk,

http://www.le.ac.uk/personnel/jobs

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

From: Greg Doherty <gdz@ansto.gov.au>

Date: Sun, 5 Aug 2001 13:00:24 +1000 (EST)

**Subject: Postdoctoral Position at Australian Nuclear Science Organisation**

Post Doctoral Position, Environment Division

Australian Nuclear Science and Technology Organisation,

southern outskirts of Sydney, NSW, Australia

Salary: Australian dollars $45,000 - $50,000; Two Year Term

PhD graduate in mathematics, geochemistry, geophysics or computing to work

in the Sulfide Solutions Research Project involving the analysis, modelling

and prediction of the time dependent behaviour of chemical and physical

transport processes in waste rock and stockpiles containing sulfidic

material. The successful applicant will provide specialised computational

code development support to a large group specialising in research into acid

mine drainage. Their work is highly regarded within Australia and the

international mining community. Client companies are located in many

countries of the world with significant market expansion predicted over the

next 2-3 years. In addition to a relevant PhD, any experience in numerical

simulation of geochemical or physical processes will be highly regarded.

Further information, including selection criteria to be addressed, can be

obtained from the ANSTO homepage, http://www.ansto.gov.au

Applications close 31 August.

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

From: Vladik Kreinovich <vladik@cs.utep.edu>

Date: Mon, 30 Jul 2001 16:22:53 -0600 (MDT)

**Subject: Contents, Reliable Computing**

6-2001

Reliable Computing

Volume 7, issue 6, 2001

Mathematical Research

Linear Interval Tolerance Problem and Linear Programming Techniques

Oliver Beaumont, Bernard Philippe

433-447

An Effective High-Order Interval Method for Validating Existence and

Uniqueness of the Solution of an IVP for an ODE

Nedialko S. Nedialkov, Kenneth R. Jackson, John D. Pryce

449-465

On the Combination of Interval Constraint Solvers

Laurent Granvilliers

467-483

The Preliminary Enclosing of the ODE Solutions on the Base of the

Cauchy-Duhamel Identity

Gregory G. Menshikov

485-495

A Surprising Approach in Interval Global Optimization

Sergey P. Shary

497-505

Information

Gregory G. Menshikov has turned 70

507-508

SCAN 2002

509-510

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

From: Maya Neytcheva <neytchev@sci.kun.nl>

Date: Wed, 01 Aug 2001 18:38:34 +0200

**Subject: Contents, Numerical Linear Algebra with Applications**

CONTENTS

Numerical Linear Algebra with Applications

Volume 8, Issues 4-5, 2001

Local preconditioners for two-level non-overlapping domain

decomposition methods

L.M. Carvalho, L. Giraud, G. Meurant (pp.207-227)

A nearly optimal preconditioner for the Navier-Stokes equations

L. Hemmingsson-Franden, A. Wathen (pp. 229-243)

Lanczos, Householder transformations, and Implicit Deflation

for fast and reliable dominant singular subspace computation

R. Fierro (pp. 245-264)

Error norm estimation and stopping criteria in

preconditioned conjugate gradient iterations

O.Axelsson and I. Kaporin (pp. 265-286)

A divide and conquer approach to computing the mean first passage

matrix for Markov chains via Perron complement reductions

S.J. Kirkland, M. Neumann, J. Xu (pp. 287-295)

A Parallel Balanced Method for Sparse Systems

G. Golub, A. Sameh, V. Sarin (pp. 297-316)

Continuation of invariant subspaces

L. Dieci, J. Friedman (pp. 317-327)

Barrier variational generation of quasi-isometric grids

V. Garanzha (pp. 329-353

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

From: George Anastassiou <anastasg@msci.memphis.edu>

Date: Thu, 02 Aug 2001 09:27:56 -0500

**Subject: Contents, Journal of Computational Analysis and Applications**

Contents Journal of Computational Analysis and Applications

Vol3, No3, 2001 (Kluwer/Plenum)

1)Effect of a localised random forcing term on the Korteweg-de Vries

equation,by Arnaud Debussche,Jacques Printems... 183

2)The central approximation theorems for the method of left Gamma

quasi-interpolants in Lp-spaces,by M.W.Muller.... 207

3)On the Distribution of Leja-Gorski points,by Mario

Gotz......................... 223

4)On the extended reversed Meir inequality,by

B.Guljas,C.E.M.Pearce,J.Pecaric....................243

5)On Global smoothness preservation by Bernstein operators,by Ioan

Gavrea.......................... 249

6)A selfconsistent parametric inference approach on construction

heuristics,by Bardo E.J.Bodmann,Arthur T.Gomez.....259

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

End of NA Digest

**************************

-------