Send mail to:    mgnet@cs.yale.edu             for the digests or bakeoff
                  mgnet-requests@cs.yale.edu    for comments or help
 Current editor:  Craig Douglas                 douglas-craig@cs.yale.edu
Anonymous ftp repository:    casper.cs.yale.edu (128.36.12.1)

Today's editor:  Craig Douglas (douglas-craig@cs.yale.edu)

Volume 3, Number 12 (December 31, 1993)

Today's topics:

     References from Ruede's Book
     Special Issues of Advances in Computational Mathematics
     Papers for MGNet
     8-th International Conference on Domain Decomposition
     7-th Domain Decomposition Papers on MGNet
     Book Announcement
     Sixth Copper Mountain Proceedings

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

Date:   Mon, 6 Dec 1993 20:43:41 +0100
From: Ulrich Ruede 
Subject: References

I have downloaded the references of my book

    "Mathematical and Computational Techniques for
        Multilevel Adaptive Methods"

to mgnet. These references have been edited by SIAM and should
therefore be bibliographically correct. Thanks to SIAM for permitting
the electronic distribution of this information!

Uli Ruede

    Editor's Note: in mgnet/papers/Ruede/mctmam-refs.bib
    -------------

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

From: Daniel Baltzer 
Date: Tue, 7 Dec 1993 15:51:41 +0100
Subject: Special Issues of Advances in Computational Mathematics

Call for Papers:

The journal Advances in Computational Mathematics would like to announce a
special issue which is to be concerned with practical and theoretical
aspects of multiscale techniques.

Advances in Computational Mathematics is a new journal devoted to all
aspects of computational mathematics, including basic theory, applications,
algorithms and software. The journal is published by Baltzer Publishing
from Basel, Switzerland, and currently has more than forty members on its
editorial board, representing all areas of computational mathematics. John
Mason of the Royal Military College of Science, Shrivenham, UK, and Charles
A. Micchelli of T.J. Watson IBM Research Center, Yorktown Heights, NY, USA
serve as editors-in-chief.

Multigrid algorithms and multilevel methods are very prominent and perhaps
the most important recently developed concepts in a central area of
computational mathematics, namely the numerical treatment of partial
differential equations. The concept of wavelets arose from rather different
sources but turns out to be closely related. In fact, while wavelets have
found their primary applications in signal processing, image analysis and
data compression, recent investigations indicate their promising potential
for the numerical treatment of operator equations as well.
Summarizing all these techniques under the term multiscale techniques is
perhaps justified by their many common features. Their enormous overall
impact on computational mathematics as well as perpetually  fruitful
feedback between applications, theory and development of tools make it an
excellent topic for a special issue of a journal with the above profile
which we hope could help synthesizing the ideas, aspects and directions of
research.

Therefore we cordially invite you to submit high quality, unpublished
manuscripts on any mathematical or algorithmic aspect of multiscale
techniques and their applications. We particularly welcome manuscripts on
wavelet and multiscale techniques for the numerical treatment of operator
equations, fast numerical linear algebra, inverse problems and statistics
or applications to geometric modelling.

All submitted papers will undergo the usual process of peer review. We hope
to publish a large number of submissions and will strive to ensure that
each paper receives careful and prompt consideration.

If you would like to submit a paper to this special issue please send three
copies of the manuscript to:

Prof. Dr. Wolfgang Dahmen,
Guest Editor, Special Issue on Multiscale Techniques,
Institut fur Geometrie und Praktische Mathematik,
RWTH Aachen,
Templergraben 55,
52056 Aachen,
Germany.
E-mail: dahmen@igpm.rwth-aachen.de

The deadline for submissions is June 30, 1994. We anticipate that the
special issue will appear early in 1995. A free sample copy of Advances in
Computational Mathematics containing 'instructions to authors' is available
from publish@baltzer.nl

Sincerely,

Wolfgang Dahmen
J.C. Baltzer AG, Science Publishers
Asterweg 1A
1031 HL Amsterdam
The Netherlands
tel. +31-20-637 0061
fax. +31-20-632 3651
e-mail: publish@baltzer.nl

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

Date: Sun, 12 Dec 1993 23:06:56 -0500
From: Jinchao Xu 
Subject: 8-th International Conference on Domain Decomposition

The Eighth International Conference on Domain Decomposition Methods
will be held in Beijing, China during May 15-19, 1995. For further
information, please contact:

    Prof. Zhong-Ci SHI
    Computing Center
    Academia Sinica
    PO Box 2719
    Beijing 100080, CHINA
    Email: zcshi%bepc2@scs.slac.stanford.edu

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

Date: Mon, 13 Dec 1993 10:58:21 -0500
From: douglas-craig (Craig Douglas)
Subject: 7-th Domain Decomposition Papers on MGNet

To put a paper and abstract on MGNet for the Seventh Domain Decomposition
Symposium, please do the following if possible (e-mail the files to the
address below if you cannot do this):

    % ftp casper.cs.yale.edu     => Initiate connection
    User: anonymous              => You type anonymous
    Password:                    => You type your e-mail address
    ftp> cd incoming             => Change to the incoming directory
    ftp> mkdir your_name         => Make a directory for yourself
    ftp> cd your_name            => Change to your directory
    ftp> hash                    => Print # symbols during transfers
    ftp> put paper.abs           => Put an abstract file (plain text, please)
    ftp> binary                  => Binary transfer
    ftp> put paper.ps            => Put the PostScript paper
    ftp> quit                    => All done

While a PostScript file is preferred, almost any type will be accepted.  The
binary transfer is very important as the ftp server on casper is very
particular about the eighth bit.  I suggest keeping the following line in the
.netrc file in your home directory:

machine casper.cs.yale.edu login anonymous password your@email.address

The e-mail address is required and casper checks if it is correct before
logging all of the transactions and giving a user access.  For people without
casper's netword address, it is 128.36.12.1.

All incoming files with be moved by me to the directory mgnet/DDM7.  The
README.mgnet file in that directory contains information about the files.

Craig Douglas
douglas-craig@cs.yale.edu

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

Date: Wed, 15 Dec 1993 09:29:08 GMT
From: Wolfgang.Joppich@gmd.de (Wolfgang Joppich)
Subject:  Book Announcement

Dear editor,

although my book has been published some months ago (September 1993)
I feel that it is appropriate to make it known to the MG-community.
Unless the book has not yet been announced by Slobodan Mijalkovic I
would ask for the following announcement in your mgnet-news.

Thank you in advance

             Wolfgang Joppich
                      
Multigrid Methods for Process Simulation

by
W. Joppich and S. Mijalkovic

Gesellschaft fuer Mathematik und Datenverarbeitung, Sankt Augustin, Germany
and
University of Nis, Nis, Yugoslavia
                  
This book combines both research in multigrid methods and
a particular application field, here -- process simulation. 
It is the declared intention of this book to 
convince by practically demonstrating the power of the
multigrid principle and to establish an example of fruitful
interdisciplinary interaction.
The introduction to multigrid is therefore strictly directed towards
the goal to provide the algorithmical overview one needs to compose             
optimal multigrid algorithms for evolution problems of process 
simulation and similar applications.
The necessary explanation how and why multigrid works is derived
from the roots. So the book preassumes no advanced familiarity
with numerical analysis. Additionally a complete strategy 
to implement different                                     
algorithmical components on an adaptive multilevel grid structure
is presented. The outlined principle of grid refinement and
adaptation is based on the control of errors and is reliable
as well as general.
Last but not least the described strategies are applied to ``real life''
problems of process simulation.
                                                                                
Consequently this book is an important contribution to the 
interdisciplinary challenge of improving numerical techniques for 
diffusion problems of process simulation.
                                                                                
Contents: 
Model problems; The multigrid principle; 
The components of the algorithm; The full approximation scheme; 
Full multigrid; Multigrid methods for refined grids;
Parabolic problems; Systems; Tools to estimate multigrid convergence;
Multigrid on parallel computers; Standard multigrid for semiconductor
device equations; Grid selection strategies; Discretization errors;
Extrapolation techniques;Local refinement criteria;
Two-level time-stepping schemes; Tayloring multigrid components
for a diffusion model problem; Results of smoothing analysis;
Prolongation of grid functions; Procedures for adaptive multigrid;
Practical programming techniques;
Adaptive transient simulation; Multiparticle evolution processes;
BiCMOS technology--Case study


Authors    : W. Joppich, S. Mijalkovic
Title      : Multigrid Methods for Process Simulation
Publisher  : Springer Wien / New York
Series     : Computational Microelectronics
Year       : 1993
ISBN       : 3-211-82404-9 Wien/New York
ISBN       : 0-387-82404-9 New York/Wien
Figures    : 126
Pages      : 309

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

Date: Thu, 16 Dec 1993 11:56:40 -0700
From: jmandel@tiger.denver.colorado.edu (Jan Mandel)
Subject: Papers for MGNet

@techreport{LeTallecMandelVidrascu,
        author="Patrick {Le Tallec} and Jan Mandel and Marina Vidrascu",
        title="Balancing Domain Decomposition for Plates",
        year="1993",
        month="December",
        type="UCD/CCM Report",
        institution="Conter for Computational Mathematics,
        University of Colorado at Denver",
        number="7",
        file="plate.ps",
        note="Proceedings of the 7th International Symposium on
        Domain Decomposition Methods, Penn State, November 1993, submitted"
        }

    Editor's Note: in mgnet/DDM7/LeTallec-Mandel-Vidrascu.ps
    -------------

@techreport{CaiMandelMcCormick-93,
        title="Multigrid Methods for Nearly Singular Linear Equations
        and Eigenvalue Problems",
        author="Zhiqiang Cai and Jan Mandel and Steve McCormick",
        year="1993",
        month="December",
        number="6",
        type="UCD/CCM Report",
        institution="Conter for Computational Mathematics,
        University of Colorado at Denver",
        file="eig-dec93.ps"
        }

    Editor's Note: in mgnet/papers/CaiZ-Mandel-McCormick/eig-dec93.ps
    -------------

@unpublished{GhosalMandelTezaur,
        authors="S. Ghosal and J. Mandel and R. Tezaur",
        title="Automatic Substructuring for Domain Decomposition
        Using Neural Networks",
        year="1993",
        month="December",
        number="5",
        type="UCD/CCM Report",
        note="IEEE Congress on Neural Nets, 1994",
        file="ghosal.ps"
        }

    Editor's Note: in mgnet/papers/Ghosal-Mandel-Tezaur/ghosal.ps
    -------------  (this is about 9.15Mb).
                    Also, Jan contributed some other papers at the same time.
                    They are in mgnet/papers/Mandel/{bdd.ps,ibim.ps}.

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

Date: Mon, 20 Dec 1993 07:18:47 -0500
From: Postman@CosCob.CT.USMail.Gov
Subject: Sixth Copper Mountain Proceedings

Hi,

I am your postman.  I have a lot of copies of a NASA report which I have to
carry to your house along with your Christmas cards, boxes, and catalogs.  Do
you mind if I wait until after the 25th?  They are bulky and kind of heavy.
Next time, please ask them to mail these things a different month (August is a
good month; I am on vacation then).

Thanks!

    Editor's Note: Thanks Duane for the proceedings.  They are mostly here by
    -------------  the 31st.  This is the weirdest e-mail message to MGNet yet.

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

End of MGNet Digest
**************************