Send mail to:             for the digests or bakeoff
         for comments or help
Anonymous ftp repository: (
Current editor:  Craig Douglas

WWW Sites: or

Today's editor:  Craig Douglas (

Volume 13, Number 12 (approximately December 31, 2003)
Volume 14, Number 1  (approximately January 31, 2004)

Today's topics:

     Important Dates
     Post doc posting (University of Utah)
     Postdoctoral Position at UCLA
     Faculty Position in Math/Comp Finance at Kent State University
     Copper Mountain Conference on Iterative Methods Papers/Extended Abstracts
        L. Giraud, D. Ruiz, A. Touhami
        H. Guo and R. A. Renaut
     WiR Seminar High Dimensional Problems
     High Performance Computing and Applications (HPCA 2004)
     Table of Contents, ETNA vol 16, 2003


Date: Sat, 31 Jan 2004 10:22:21 -0400
From: Craig Douglas 
Subject: Important Dates

February 1:  Author abstracts due for the Copper Mountain Conference on
Iterative Methods.  See
Note also for hotel and early registration (February 27).

March 1:  Author abstracts due for WiR Seminar on Efficient Numerical Methods
for High Dimensional Problems.  Send 200 word abstract (pdf format) to


Date: Sun, 14 Dec 2003 18:19:42 -0700 (MST)
From: Martin Berzins 
Subject: Post doc posting (University of Utah)

Postdoctoral Researcher in Adaptive Mesh Refinement at University of Utah

A postdoctoral researcher is required to work on adaptive mesh refinement
(AMR) algorithms for The Center for the Simulation of Accidental Fires and
Explosions (C-SAFE), a DOE ASCI ASAP Level 1 center at the University of Utah
in Salt Lake City.

This position requires experience in numerical algorithm development and
extensive programming expertise in C++.  Specific experience with cartesion
mesh refinement (Structured AMR), while not required, is highly desirable.
The applicant should have a solid mathematics background and some experience
with CFD algorithms (preferably multiprocessor/parallel codes).  Experience
with Lagrangian particle simulations is also valuable, but not necessary.

The research aspects of the project will involve developing novel mesh
refinement algorithms for combined fixed mesh and particle codes being used
for modeling complex transient reacting fluid-structure interactions.
Development aspects include helping develop and extend the existing parallel
MPI C++ software framework and integrating new AMR algorithms into the
framework.  The applicant will be ultimately responsible for this portion of
the software development.

Ability to work as part of a large multidisciplinary team is essential.  The
applicant will work closely with chemical and mechanical engineers in
developing AMR algorithms for combustion and structural mechanics, and with
computer scientists in the development of the AMR S/W framework.

C-SAFE Overview:  C-SAFE focuses specifically on providing state-of-the-art,
science-based tools for the numerical simulation of accidental fires and
explosions, especially within the context of handling and storage of highly
flammable materials.  The objective of C-SAFE is to provide a system
comprising a problem-solving environment in which fundamental chemistry and
engineering physics are fully coupled with non-linear solvers, optimization,
computational steering, visualization and experimental data verification.
More details may be found at

This position is with the University of Utah's Scientific Computing and
Imaging (SCI) Institute (  The applicant will work
with Dr. Martin Berzins and Dr. Steve Parker and with a multidisciplinary
team of faculty, staff, and students.

Please apply to Dr. Greg Jones (jones@sci.utah.ed).  Projected closing date
for applications is Jan 31, 2004.

Salary Range:  50 - 60K (Depending on experience.)
Benefits:      The University of Utah offers a very competitive
               benefits package.  See
Start date:    Feb, 2004. 


Date: Mon, 15 Dec 2003 07:28:17 +0200
From: Achi Brandt 
Subject: Postdoctoral Position at UCLA


This is a call for applications to a postdoctoral position at UCLA involving a
well defined research in scientific computation.

The research, entitled "Numerical Coarsening and Upscaling for Geophysical
Fluid Dynamics", is led by Prof. Achi Brandt, in the Department of
Mathematics, and by Prof. James C. McWilliams in the Department of
Atmospheric and Oceanic Sciences.

A full technical description of the project can be viewed in (McW.pdf).  In essence, the project
involves the development of new, highly innovative methods of scientific
computation.  At the minimum it requires background in numerical solution of
partial differential equations.  Very desired is knowledge and experience more
specifically in computational fluid dynamics and/or in multiscale (e.g.,
multigrid) methods.

The research project is part of a broader activity aimed at the development of
multiscale methods in general (see a review in (review.pdf)) and the methodology of
systematic upscaling in particular (see
(UPSCALE-B1.pdf)).  These modern approaches have far reaching implications in
most fields of scientific computation, so the involved researcher can expect
to gain highly valuable advanced knowledge and experience.

Interested candidates should write to both and  In your letter please state what is the background that
makes you capable and interested in this project.

    Achi Brandt
    Jim McWilliams


Date: Fri, 9 Jan 2004 21:02:27 -0500
From: Chuck Gartland 
Subject: Faculty Position in Math/Comp Finance at Kent State University

Kent State University
Department of Mathematical Sciences
Kent, OH 44242

Tenure-Track Positions in Mathematics

We invite applications for one tenure-track position in applied
mathematics in the areas of financial mathematics or computational
finance and one or more tenure-track positions (pending budget
approval) in mathematics in the areas of algebra or analysis. The
appointments are to be at the Assistant Professor level and are to
begin August 16, 2004.

Candidates are required to have a Ph.D. within the mathematical
sciences and preference will be given to candidates with some
postdoctoral experience. They are expected to have strong potential in
research (including the potential to attract external funding) and in
teaching. They should be able to contribute to the interdisciplinary
outreach of the department or to support established research

The successful candidate for the applied mathematics position will be
expected to contribute to and help develop our highly-rated,
cross-disciplinary Master's program in Financial Engineering
( and should have a background in
financial mathematics and/or computational finance. Desirable research
areas include, but are not limited to, theoretical or computational
aspects of differential equations (stochastic or deterministic),
statistical modeling, optimization, and wavelet analysis.

Candidates in algebra or analysis should have a background that
complements or broadens our existing strengths in these areas.

Kent State University is a spacious, residential campus serving more
than 24,000 students.  It is situated in a small university town
within 30 miles of the major metropolitan area of Cleveland, Ohio. The
Department of Mathematical Sciences is in the College of Arts and
Sciences and houses programs through the doctoral level in pure and
applied mathematics.  There are currently 24 tenure-track or tenured
faculty. For further information about the department, please visit
the website (

Applicants should send a cover letter, a curriculum vitae, and at
least three letters of reference to the Mathematics Search Committee
at the above address. Applicants are also requested to use the AMS
standardized application format, forms for which are available through
the American Mathematical Society ( Questions
regarding these positions may be sent to Applicants whose completed applications are
received by January 20, 2004, are assured of receiving full

Kent State University is an Equal Opportunity, Affirmative Action Employer.


Date: Thu, 15 Jan 2004 13:32:53 +0100
From: Ahmed Touhami 
Subject: Copper 2004 Expanded Abstract 

A comparative study of iterative solvers exploiting spectral information
for SPD systems

L. Giraud, D. Ruiz, A. Touhami



We intend to solve the linear system Ax= b where A is large sparse and
symmetric positive definite. We consider solution techniques that
exploit spectral information about the matrix A. In this talk, we will
describe the various techniques as well as the observed numerical
behaviours on a set of model problems from Matrix Market. We will
discuss their numerical efficiency, computational complexity and
sensitivity to the accuracy of the spectral information.

Keywords: Chebyshev polynomial, Block Lanczos method, filtering,
deflation, preconditioning, iterative methods.

    Editor's Note: see

Name: Ahmed Touhami
Address: ENSEEIHT-IRIT, 2 rue Charles Camichel, 31071 Toulouse Cedex, France.
Phone: + 33 (0) 5 61 58 80 49
Fax:   + 33 (0) 5 61 58 83 06


Date: Tue, 06 Jan 2004 14:58:40 -0700
From: Rosemary Renaut 
Subject: Copper 2004 Paper

Parallel Variable Distribution for Total Least Squares

Hongbin Guo and Rosemary A. Renaut

Department of Mathematics and Statistics
Arizona State University
Tempe, AZ 85287 1804
{hb_guo, renaut}


A novel parallel method for determining an approximate total least squares
(TLS) solution is introduced.  Based on domain distribution, the global TLS
problem is partitioned into several dependent TLS subproblems.  A convergent
algorithm using the parallel variable distribution technique (Ferris and
Mangasarian, 1994) is presented.  Numerical results support the development
and analysis of the algorithms.

    Editor's Note: see


Date: Tue, 23 Dec 2003 10:27:03 +0100
From: Jochen Hittler 
Subject: WiR Seminar High Dimensional Problems

                                WiR Seminar on
                       Efficient Numerical Methods for
                          High Dimensional Problems

                           Hohenwart Forum, Germany
                              May 16 - 19, 2004

        GAMM Fachausschuss Scientific Computing, Research Network WiR
   Baden-Wuerttemberg, University of Heidelberg / Simulation in Technology

Objectives and scopes

High-dimensional problems are one of the great challenges to the future
development of efficient numerical methods.  Nowadays Monte Carlo methods are
used for such problems.  For certain problem classes special efficient
numerical methods have been developed which can be discussed on this Seminar.

Topics are
- High dimensional partial differential equations, integral equations and
- Problems with high dimensions from Financial Mathematics
- Schroedinger equation, high dimensional phase space discretizations,
  Boltzmann equation, path integral discretzations
- Combustion problems, pdf-methods, ILDM
- Sparse grid methods, dimensional reduction, quasi Monte Carlo methods,
  stochastic solution methods, particle methods


The conference will be held at Pforzheim-Hohenwart, Germany at the Hohenwart
Forum Conference Center.  Hohenwart Forum is situated in thern Northern Black
Forest.  It is about 1 hour to travel from Franfurt/Main Airport.

Paper submission

Authors should submit abstracts of about 200 words as pdf-file to  Notice of acceptance will be given up to March 15,
2004.  All accepted papers will be published in a volume available at the time
of the conference.


March 1, 2004: Abstracts due
April 1, 2004: Registration due
March 15, 2004: Notification of acceptance


M. Griebel, Bonn, W. Hackbusch, Leipzig, U. Maas, Stuttgart, G. Wittum,


All information on the workshop will be available at .
Registration office:  Phone +49 (0) 6221 5488-66 or -54,
Fax:  +49 (0) 6221 5488-60,


Date: Wed, 14 Jan 2004 15:47:12 -0600
From: "Chen, Zhangxin" 
Subject: High Performance Computing and Applications (HPCA 2004)

Announcement and Call for Papers and Organizing Minisymposia

     High Performance Computing and Applications (HPCA 2004)
                  August 8--10, 2004
           Shanghai University, Shanghai, China

* China Computer Federation
* National Natural Science Foundation of China
* Shanghai Computer Society
* Shanghai University
* Xi'an Jiaotong University

Advisory Committee:
Jialin Cao (Chair), Vice President of Shanghai University, China
James G. Glimm, State University of New York at Stony Brook, USA
David Keyes, Columbia University, USA
Sanli Li, Shanghai University and Tsinghua University
Jiachang Sun, Institute of Software, Academia, China
Wei Zhao, Texas A&M University, USA

Scientific Committee:
Shiyi Chen, The Johns Hopkins University, USA
Yaosong Chen, Peking University, China
Jim Douglas, Jr., Purdue University, USA
Richard E. Ewing, Texas A&M University, USA
Roland Glowinski (Chair), University of Houston, USA
Vipin Kumar, University of Minnesota, USA
Do Y. Kwak, KAIST, South Korea
Hermilo R. Leon, Mexican Institute of Petroleum, Mexico
Kaitai Li, Xi'an Jiaotong University, China
Xinda Lu, Shanghai Jiaotong University, China
Yuanle Ma, Tsinghua University, China
Peter Moore, Southern Methodist University, USA
Depei Qian, Xi'an Jiaotong University, China
Gengfeng Wu, Shanghai University, China

Invited Speakers:
Wei Cai, University of North Caroline, Charlotte, USA
Shiyi Chen, John Hopkins University, USA
Zhangxin Chen, Southern Methodist University, USA
Zhiming Chen, Chinese Academy of Sciences, China
Jim Douglas, Jr., Purdue University, USA
Craig Douglas, University of Kentucky and Yale University, USA
Weinan E, Princeton University, USA
Richard E. Ewing, Texas A&M University, USA
Roland Glowinski, University of Houston, USA
Jiwen He, University of Houston, USA
Song Jiang, IAPCM, China
Kaitai Li, Xi'an Jiaotong University, China
Xiaolin Li, SUNY at Stony Brook, USA
Tsorng-Whay Pan, University of Houston, USA
Yuehong Qian , Princeton University, USA
Thomas Russell, University of Colorado at Denver, USA
Jiachang Sun, Institute of Software,Academia Sinica, China
Junping Wang, Colorado School of Mines, USA
Nong Xiao, National University of Defense Technology, China
Pingwen Zhang, Peking University, China
Wu Zhang, Shanghai University, China
Jun Zou, Chinese University of Hong Kong, China

Conference Themes:
(1) Numerical Algorithms and Solutions:  Numerical ODEs & PDEs, numerical
integral equations, finite difference, volume, and element methods, boundary
element methods, Monte Carlo methods, numerical linear algebra, multiscale
methods, and treatment of uncertainty.
(2) Parallel Computing:  Parallel solvers, multigrid and multilevel methods,
domain decomposition methods, and parallel computing strategies.
(3) Novel High Performance Computing Architecture and Platform General
architectures for high performance computing systems, toolkits, development
environments, protocols, and languages, performance evaluation models and
tools, and QoS models and tools.
(4) Applications:  Scientific computing for physical problems, computational
fluid dynamics mathematical modeling, numerical simulation, engineering
applications, computer visualization, and industrial applications

Conference Format:
The conference will consist of 45-minute invited talks and 30-minute
minisymposia contributed talks.  The conference language is English.

Conference Proceedings:
Proceedings of presented and refereed papers will be published.

Call for Papers and Organizing Minisymposia:
Participants wishing to organize a minisymposium or present a paper in a field
related to the conference themes are invited to submit a one-page abstract in
English to  Please include the address, telephone,
fax, and e-mail address of the primary contact person.


Date: Mon, 19 Jan 2004 09:39:45 -0500 (EST)
From: Lothar Reichel 
Subject: Table of Contents, ETNA vol 16, 2003

Table of Contents, Electronic Transactions on Numerical Analysis (ETNA),
vol. 16, 2003. ETNA is available at and at several 
mirror sites, as well as on CDROM. 

ETNA is in the extended Science Citation Index and the CompuMath Citation 

Stefano Serra Capizzano and Cristina Tablino Possio, Preconditioning 
strategies for 2D finite difference matrix sequences, pp. 1-29.

Doron Chen and Sivan Toledo, Vaidya's preconditioners: implementation and 
experimental study, pp. 30-49.

Henri Schurz, General theorems for numerical approximation of stochastic 
processes on the Hilbert Space, pp. 50-69.

Martin Bohme and Daniel Potts, A fast algorithm for filtering and wavelet 
decomposition on the sphere, pp. 70-92.

Heinrich Voss, A rational spectral problem in fluid-solid vibration, 
pp. 93-106.

Limin Wu, A parameter choice method for Tikhonov regularization, pp. 107-128.

A. El Guennouni, K. Jbilou, and H. Sadok, A block version of BiCGStab for
linear systems with multiple right-hand sides, pp. 129-142.

J. Illa'n, A quadrature formula of rational type for integrands with one
endpoint singularity, pp. 143-164.

S. C. Brenner, Analysis of two-dimensional FETI-DP preconditioners by the
standard additive Schwarz framework, pp. 165-185.

F. Luengo and M. Raydan, Gradient method with dynamical retards for
large-scale optimization problems, pp. 186-193.


End of MGNet Digest