Send mail to: mgnet@cs.yale.edu for the digests or bakeoff
mgnet-requests@cs.yale.edu for comments or help
Anonymous ftp repository: www.mgnet.org (128.163.209.19)
Current editor: Craig Douglas douglas-craig@cs.yale.edu
World Wide Web: http://www.mgnet.org or
http://casper.cs.yale.edu/mgnet/www/mgnet.html or
http://www.cerfacs.fr/~douglas/mgnet.html or
http://phase.etl.go.jp/mgnet or
http://www.nchc.gov.tw/RESEARCH/Math/mgnet/www/mgnet.html
Today's editor: Craig Douglas (douglas-craig@cs.yale.edu)
Volume 10, Number 3 (approximately March 31, 2000)
Today's topics:
Copper Mountain 2000
Postdoctoral position at the Weizmann Institute
Postdoctoral Position Available (Rutgers)
Scientific Computing: Reader/Professor, Lecturers
Summer School/Workshop Announcement (Penn State)
Numerical Analysis Summer School at University of Durham
New Publication (Ovtchinnikov and Xanthis)
New Report (Benzi, Nabben, and Szyld)
Copper Mountain Paper (Vuik, Segal, Meijerink, and Frank)
Copper Mountain Paper (Baryamureeba)
F 95 sparse BLAS
-------------------------------------------------------
Date: Fri, 31 Mar 2000 10:15:12 +0500
From: Craig Douglas
Subject: Copper Mountain 2000
Anyone who is attending the Iterative Methods conference at Copper Mountain
(April 3-7) is invited to put a paper in the virtual proceedings. Please
contact me directly or send your paper as an email attachment to
mgnet@cs.yale.edu. The papers will be collected in
http://www.mgnet.org/mgnet-cm2000.html.
-------------------------------------------------------
Date: Wed, 8 Mar 2000 14:33:59 +0200 (IST)
From: Carol Weintraub
Subject: Postdoctoral position at the Weizmann Institute
The Gauss Center for Scientific Computation at the Weizmann Institute
of Science, Rehovot, Israel, is seeking a qualified candidate for a
Post-Doctoral position, to participate in the Center's development of
advanced multiscale computational methods. Possible directions include:
1. Multigrid methods, especially in fluid dynamics
2. Data assimilation in atmospheric simulations
3. Ab-initio quantum chemistry (many-eigenfunction problems)
4. Wave phenomena
5. Optimal feedback control
6. Computational physics: multiscale Monte-Carlo and Dirac solvers
7. Fast medical-imaging and radar reconstruction
8. Molecular dynamics
9. Multiscale image processing
Applicants should have completed their Ph.d. studies between
June 1998 and the appointment start date. They should have a
strong background and programming experience in scientific computing,
preferably in one of the above areas and/or in multiscale computational
methods.
The post-doc appointment is for a period of one year and in some cases
may be renewed for a second year.
Applications and request for further information should be sent to
the Gauss Center Secretary:
Mrs. Carol Weintraub
Dept. of Computer Science & Applied Mathematics
The Weizmann Institute of Science
Rehovot 76100, Israel
Email: carol@wisdom.weizmann.ac.il
Fax: +972-8-9342945
Phone: +972-8-9343545
The application should include a CV, a description of previous
experience in scientific computing and the particular areas of
interest, and names of two referees, who should separately send
their letters of recommendation to the above address.
More information on the research at the Gauss Center can be found
at our web site:
http://www.wisdom.weizmann.ac.il/~achi/gaussctr.html
-------------------------------------------------------
Date: Fri, 24 Mar 2000 16:20:10 -0500 (EST)
From: knight@donizetti.rutgers.edu
Subject: Postdoctoral Position Available (Rutgers)
Postdoctoral Fellow (Research Associate)
Computational Fluid Dynamics
Department of Mechanical and Aerospace Engineering
Rutgers University
Piscataway, New Jersey USA
A postdoctoral position in computational fluid dynamics is available effective
immediately in the Department of Mechanical and Aerospace Engineering at
Rutgers University. The individual will participate in the development and
application of Large Eddy Simulation (LES) methodology. Qualifications
include a PhD in Mechanical or Aerospace Engineering, experience in numerical
simulation using LES and/or the Reynolds-averaged Navier-Stokes equations, and
knowledge of reacting flows. Initial appointment will be for one year, with
the possible renewal of an additional year. Interested persons should submit
a resume and the names of three references to:
Professor Doyle Knight
Department of Mechanical and Aerospace Engineering
Rutgers University-The State University of New Jersey
98 Brett Road
Piscataway, NJ 08854-8058
Telephone: 732 445 4464
Faxsimile: 732 445 5313
Email: knight@jove.rutgers.edu
Website: http://www-srac.rutgers.edu/~knight
-------------------------------------------------------
Date: Mon, 6 Mar 2000 14:41:26 GMT
From: Leonidas Xanthis
Subject: Scientific Computing: Reader/Professor, Lecturers
The University of Westminster (Harrow School of Computer Science)
is advertising (CLOSING date: 13th March, 2000)
1 x Readership/Professorship
2 x Lectureships/Senior Lectureships
0.5 Lectureship
for teaching and research in the various areas taught at the
School of Computer Science at its Harrow Campus.
One of the areas is
SCIENTIFIC COMPUTING with emphasis in:
1. applied mathematics-mechanics-materials
2. iterative methods for large-scale problems
3. 3D visualisation and modelling-computer graphics-mathematics
and finite element modelling in solids
Prospective applicants with outstanding research record in
Scientific Computing are welcome to contact informally
Leonidas Xanthis
Centre for Techno-Mathematics &
Scientific Computing Laboratory,
University of Westminster,
LONDON HA1 3TP, UK
FAX: 00-44-171-911 5906
-------------------------------------------------------
Date: Mon, 20 Mar 2000 01:02:46 -0500 (EST)
From: Ludmil Tomov Zikatanov
Subject: Summer School/Workshop Announcement (Penn State)
Advanced Summer School/Workshop on Multigrid Methods
An advanced summer school/workshop on multigrid methods will be held in the
Center for Computational Mathematics and Applications at the Pennsylvania
State University during June 19-24, 2000. The principal speakers/organizers
will be Professors Achi Brandt and Jinchao Xu, and several other active
multigrid researchers will also participate. The emphasis of the summer
school is on the application of multigrid methods to practical problems
including Maxwell and Helmholtz equations, Navier-Stokes equations and
ab-initio quantum physics/chemistry.
More information can be found on the following URL address:
http://www.math.psu.edu/xu/MGworkshop/
Ludmil Zikatanov
Department of Mathematics Office: (814) 865-3517
208 McAllister Building Email: ltz@math.psu.edu
Penn State University URL: http://www.math.psu.edu/ltz/
University Park, PA 16802
-------------------------------------------------------
Date: Tue, 28 Mar 2000 14:38:41 +0100
From: Alan Craig
Subject: Numerical Analysis Summer School at University of Durham
Numerical Analysis Summer School at University of Durham
(including nonlinear multigrid course)
The 9th EPSRC Summer School in Numerical Analysis will take place in the
University of Durham, United Kingdom on the 9th-21st July, 2000.
The speakers and topics will be
Sunday 9th July - Friday 14th July 2000
C. Bernardi (Paris): Spectral and High order methods.
P. Kloeden (Frankfurt): Stochastic Differential Equations and their Numerical
Solution.
A. Szepessy (Stockholm): Hyperbolic Differential Equations.
Sunday 16th July - Friday 21st July 2000
P. Bjorstad (Bergen): Domain Decomposition Techniques.
C. Carstensen (Keil): Microstructure.
R. Kornhuber (Berlin): Nonlinear Multigrid Techniques.
The purpose of the Summer Schools has always been the delivery of high quality
instructional courses at postgraduate level. The intended audience is a wide
one, and will include research students, academics and people from industry.
The lecturers will be encouraged to enable a person with an elementary
knowledge of a subject to attain reasonable competence in it, or with a good
initial knowledge to identify and begin work on unsolved problems.
Each week, thirty five funded places are available for participants eligible
for EPSRC funding. These include EPSRC sponsored post-graduate students and
UK academics. Priority is given to post-graduate students. The funding
covers the cost of accommodation, but does not include travel expenses. A
nominal registration fee is charged for EPSRC supported participants. Details
of the fees are included on the web page linked below.
Others, including those from industry and overseas, not eligible for EPSRC
support are liable for the cost of registration, accommodation and meals.
For further details (including online registration) see the web page
http://maths.dur.ac.uk/nass
or contact
Alan Craig
Department of Mathematical Sciences
University of Durham
South Road
Durham DH1 3LE
England
alan.craig@durham.ac.uk
+44 (0) 191 374 2381
-------------------------------------------------------
Date: Mon, 6 Mar 2000 14:41:26 GMT
From: Leonidas Xanthis
Subject: New Publication (Ovtchinnikov and Xanthis)
Title: Effective dimensional reduction algorithm for eigenvalue problems for
thin elastic structures: A paradigm in three dimensions
Authors: E.E. Ovtchinnikov and L.S. Xanthis
Journal: Proc. Natl. Acad. Sci. USA, February 1, 2000, vol. 97, pp. 967-971
-------------------------------------------------------
Date: Mon, 6 Mar 2000 18:54:37 -0500
From: "Daniel B. Szyld"
Subject: New Report (Benzi, Nabben, and Szyld)
Michele Benzi, Reinhard Nabben, and Daniel B. Szyld,
Algebraic Theory of Multiplicative Schwarz Methods,
Research Report 00-2-10, Department of Mathematics,
Temple University, February 2000.
Availabble at http://www.math.temple.edu/~szyld
Abstract: The convergence of multiplicative Schwarz-type methods for solving
linear systems when the coefficient matrix is either a nonsingular M-matrix or
a symmetric positive definite matrix is studied using classical and new
results from the theory of splittings. The effect on convergence of
algorithmic parameters such as the number of subdomains, the amount of
overlap, the result of inexact local solves and of "coarse grid" corrections
(global coarse solves) is analyzed in an algebraic setting. Some results on
algebraic additive Schwarz are also included.
-------------------------------------------------------
Date: Thu, 16 Mar 2000 14:18:00 +0100 (MET)
From: Kees Vuik
Subject: Copper Mountain Paper (Vuik, Segal, Meijerink, and Frank)
Deflated ICCG applied to problems with extreme contrasts in the coefficients
C. Vuik, A. Segal, J.A. Meijerink, and J. Frank
Abstract
A Deflated ICCG method is applied to problems with extreme contrasts in the
coefficients, such as those encountered in reservoir simulation, where the
permeabilities of various rocks differ orders of magnitude. The coefficient
matrix, which is symmetric and positive definite, has a very large condition
number. We prove that the the number of small eigenvalues of the IC
preconditioned matrix is equal to the number of high-permeability domains that
are not connected to a Dirichlet boundary. To annihilate the bad effect of
these small eigenvalues on the convergence, the Deflated ICCG method is used.
An efficient method is given to construct the deflation subspace, based on
physical insight, which subsequently is proven to be a good approximation of
the span of the 'small' eigenvectors. Another method is presented to
construct algebraic deflation vectors. An upperbound is given for the
resulting iteration matrix that is independent of the jumps in the
permeabilities. This choice leads to a method, which is very suitable for
parallel computing. Both choices are illustrated by numerical experiments.
Keywords: deflation, porous media, discontinuous coefficients across layers,
sensitivity analysis, incomplete factorizations, parallel computing, conjugate
gradients
AMS Subject Classification: 65F10, 65F15, 65Y05, 76S05
Editor's Note: See http://www.mgnet.org/mgnet-cm2000.html.
-------------
-------------------------------------------------------
Date: Fri, 24 Mar 2000 07:51:04 +0100 (MET)
From: Baryamureeba Venansius
Subject: Copper Mountain Paper (Venansius)
Solution of Robust Linear Regression Problems by Preconditioned Conjugate
Gradient Type Methods
Venansius Baryamureeba
March 22, 2000
Abstract: In this paper, we consider solving the robust linear regression
problem by an inexact method. We show that iteratively reweighted least
squares method and Newton method can each be combined with an iterative method
to solve large, sparse, rectangular systems of linear, algebraic equations
efficiently . We suggest and discuss preconditioners based on a new convex
weighting function. One of these preconditioners is a constant preconditioner
which is computed once at the beginning of the iterative process. For these
preconditioners, we show that the upper bound on the spectral condition number
of the preconditioned matrix is determined by a predefined constant, and is
independent of the conditioning of the data matrix. Finally we give two sets
of numerical results. One set is on demonstrating the properties of the
preconditioners based on the new convex weighting function. The second set is
a comparison with other convex weighting function on solving some given test
problems. Both sets of numerical testing clearly show that preconditioners
based on the new function are very effective in solving robust linear
regression problems.
Key words: Robust linear regression, Iteratively reweighted least squares
method, Newton's method, New weighting function, Conjugate gradient least
squares method, Preconditioner
AMS subject classification: 62J05, 65D10, 65F10, 65F20.
Editor's Note: See http://www.mgnet.org/mgnet-cm2000.html.
-------------
-------------------------------------------------------
Date: Wed, 29 Mar 2000 19:53:40 +0100 (BST)
From: Iain Duff
Subject: F 95 sparse BLAS
The current implementation and its availability is described in the CERFACS
Technical Report TR/PA/00/18 available through our Web page from the Algo home
page http://www.cerfacs.fr/algor/
Send us the bugs !!!
Editor's Note: Wow... a request for bugs. Send your favorites now :-)
-------------
-------------------------------------------------------
Date: Sun, 05 Mar 2000 18:33:55 -0400
From: "Laurence T. Yang"
Subject: CFP: HPSECA-00
CALL FOR PAPERS
Workshop on High Performance Scientific and Engineering Computing with
Applications
The Westin Harbour Castle, Toronto, Canada, August 21-24, 2000
in conjunction with
2000 INTERNATIONAL CONFERENCE ON PARALLEL PROCESSING (ICPP-2000)
(http://www.cis.ohio-state.edu/~icpp2000).
Scope and Interests:
Parallel and distributed scientific and engineering computing has become a key
technology which will play an important part in determining, or at least
shaping, future research and development activities in many academic and
industrial branches. This special workshop is to bring together computer
scientists, applied mathematicians and researchers to present, discuss and
exchange idea, results, work in progress and experience of research in the
area of parallel and distributed computing for problems in science and
engineering applications.
Among the main topics (but not limited to) are:
development of advanced parallel and distributed methods,
parallel and distributed computing techniques and codes,
practical experiences using various supercomputers with software such as MPI,
PVM, and High Performance Fortran, OpenMP, etc.
applications to numerical fluid mechanics and material sciences,
applications to signal and image processing, dynamic systems, semiconductor
technology, and electronic circuits and system design etc.
Submission Information:
Authors should send one copy of paper to the workshop organizers
(lyang@stfx.ca or pan@cps.udayton.edu) via electronic mail or three
copies via postal mail . Contributions will be reviewed for relevance and
technical contents on basis of papers. Accepted papers will be published by
IEEE Computer Society Press as proceedings of the ICPP 2000 workshops.
Further information about the conference proceedings and registration
fee can be found by web sites:
http://www.cis.ohio-state.edu/~icpp2000
http://www.stfx.ca/people/lyang/activities/icpp00-hpseca.html
Important Deadlines:
Paper submission Due April 1, 2000
Notification of Acceptance May 1, 2000
Final camera-ready paperJune 1, 2000
Workshop Organizers:
Prof. Laurence T. Yang (chair)
Department of Computer Science
PO Box 5000, St. Francis Xavier University
Antigonish, B2G 2W5, Nova Scotia, Canada
Email: lyang@stfx.ca
Prof. Yi Pan (Co-Chair)
Department of Computer Science, University of Dayton
Dayton, OH 45469-2160, USA
Email: pan@cps.udayton.edu
Technical Committee:
Hamid R. Arabnia, University of Georgia, USA
Martin Buecker, Aachen University of Technology, Germany
Jack Dongarra, University of Tennessee and
Oak Ridge National Laboratory, USA
Len Freeman, University of Manchester, UK
John Gustafson, Ames National Laboratory, USA
Sanli Li, Tsinghua University, P. R. China
Svetozar Margenov, Center Laboratory for Parallel Progressing
Bulgarian Academy of Science, Bulgaria
Russ Miller, State University of New York, Buffalo, USA
Ahmed Sameh, Purdue University, USA
Eric de Sturler, Swiss Center for Scientific Computing (SCSC-ETHZ) and
Swiss Federal Institute of Technology, The Switzerland
Karen Tomko, Wright State University, USA
Xian-He Sun, Illinois Institute of Technology, USA
Zahari Zlatev, National Environmental Research Institute, Denmark
Xiaodong Zhang, College of William and Mary , USA
------------------------------
End of MGNet Digest
**************************