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)
                             ftp.cerfacs.fr     (138.63.200.33)

World Wide Web:  http://na.cs.yale.edu/mgnet/www/mgnet.html or
                 http://www.cerfacs.fr/~douglas/mgnet.html

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

Volume 5, Number 11 (approximately November 30, 1995)

Today's topics:

     Preprints by Zhang et al
     Preprints by Pasciak et al
     Papers by Shapira
     UG Software available
     CFD short course
     Book by D'yakonov
     Faculty opening at CU-Denver
     Faculty opening at Virginia Tech
     Updated preprint by Douglas

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

Date: Mon, 20 Nov 1995 13:53:32 -0500
From: Jun Zhang 
Subject: Preprints by Zhang et al

                   An Accurate and Stable Multigrid Method 
                    for Convection-Diffusion Equations   

                Murli M. Gupta, Jules Kouatchou and Jun Zhang
                                       
                          Department of Mathematics 
                       The George Washington University
                          Washington, DC 20052, USA
                                       
                                  Abstract: 

We introduce a high-order compact difference scheme with multigrid algorithm
to solve the convection-diffusion equations with constant coefficients.  This
high-order discretization scheme is shown to be more accurate and stable than
the usual five-point discretization scheme.  It solves the convection-
diffusion equations directly without using any preconditioner or adding any
artificial dissipation terms.  This method is shown to converge faster than
some of the existing methods and to achieve higher accuracy.  Numerical
experiments are presented to validate the conclusions.

    Editor's Note: in mgnet/papers/Gupta-Kouatchou-Zhang/con_diff.ps.gz and
    -------------     mgnet/papers/Gupta-Kouatchou-Zhang/con_diff.abs.


              Acceleration of Five-point Red-Black Gauss-Seidel
                      in Multigrid for Poisson Equation

                                  Jun Zhang

                          Department of Mathematics 
                     The  George Washington University,  
                          Washington, DC 20052, USA

                                 Abstract:  

A new relaxation analysis and two acceleration schemes are proposed for the
five-point Red-Black Gauss-Seidel smoothing in multigrid for solving two
dimensional Poisson equation.  For a multigrid V cycle, we discovered that
under-relaxation is applicable to restriction half cycle and over-relaxation
is applicable to interpolation half cycle.  Numerical experiments using
modified multigrid V cycle algorithms show that our simple acceleration
schemes accelerate the convergence rate by as much as 34% with negligible
cost.  This result is contrary to the existing belief that SOR is not suitable
for using as a smoother in multigrid for Poisson equation, because the gain in
computational savings would not pay for the cost of implementing it.  More
important is the idea of employing different parameters to accelerate the
reduction of low and high frequency errors separately.  Our discovery offers a
new way for SOR smoothing in multigrid.

    Editor's Note: in mgnet/papers/Zhang/sor.ps.gz and
    -------------     mgnet/papers/Zhang/sor.abs.

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

Date: Mon, 6 Nov 1995 08:47:05 -0500
From: pasciak@jep.das.bnl.gov (Joe Pasciak)
Subject: Preprints by Pasciak et al

I put the papers and abstracts into incoming/pasciak.  Some of these papers
are not necessarily directly multigrid related but all have something to do
with preconditioning so they may be of some interest to members of the net.

Joe

    Editor's Note: I moved all 7 of the new contributions to the directory
    -------------  mgnet/papers/Pasciak-et-al.

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

Date: Tue, 14 Nov 1995 22:05:44 -0700
From: Yair Shapira 
Subject: Papers by Shapira

I put one more paper on mgnet. I also put an updated version of
the paper tlanalysis.ps.gz. Please put them in mgnet/papers/Shapira.

                         Improved Black Box Multigrid
                     for Definite and Indefinite Problems

                                 Yair Shapira
          Computer Science Department, Technion, Haifa 32000, Israel
                           e-mail: yairs@lanl.gov.

                                   Abstract

A two-level analysis method for certain separable problems is introduced.
Unlike standard two-level analysis methods, based on Fourier analysis, it is
based on spectral analysis, hence applicable to non-normal problems and to
certain problems with variable coefficients.  It motivates the definition of
improved versions of Black Box Multigrid for diffusion problems with
discontinuous coefficients and indefinite Helmholtz equations.  For
anisotropic problems, it helps in choosing suitable implementations for
frequency decomposition multigrid methods.  For highly indefinite problems, it
provides a way to choose a suitable mesh size for the coarsest grid used.
Numerical experiments confirm the analysis and show the advantage of the
present methods.

    Editor's Note: in mgnet/papers/Shapira/bbox.ps.gz and
    -------------     mgnet/papers/Shapira/bbox.abs.
                   update is mgnet/papers/Shapira/tlanalysis.ps.gz.

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

Date: Mon, 27 Nov 1995
From: Peter Bastian 
Subject: UG Software available

Dear colleagues,

we are happy to announce the availability of version 3.1 of our software "UG"
(short hand for Unstructured Grids).

UG is a flexible software library for the development of adaptive multigrid
methods on unstructured meshes. The main idea of UG was to implement most parts
in a problem-independent way, so that the software can be reused for many
different applications.

The main features of the problem-independent part are:

* 2D (triangles, quadrilaterals) and 3D (tetrahedra) unstructured meshes
* tree-based local refinement and coarsening
* user-definable number of degrees of freedom in nodes, edges, faces and
  elements (conforming, non-conforming and mixed finite-elements can be 
  implemented)
* BLAS and iterative methods (smoothers, multigrid, cg)
* powerful script language
* Multi-window graphical user interface for X11 and Macintosh. No special
  graphics libraries required
* 1000 pages of documentation (man-pages, postscript)
* Parallel version with reduced functionality available (2D, degrees of 
  freedom only in nodes, dynamic load balancing, interfaces for PVM, NX,
  PARIX, T3D, MPI)

Applications in the current distribution include:

* 2D scalar diffusion equation with linear conforming and nonconforming
  finite-elements
* 2D linear elasticity with conforming and non-conforming finite-elements
* 2D incompressible, stationary Navier-Stokes equations
* 3D convection-diffusion equation

Many other applications are currently being developped on the basis of UG.
UG is available for free for all non-profit organizations. For further 
information use:

* World Wide Web: http://www.ica3.uni-stuttgart.de/~ug
* FTP:            ftp.ica3.uni-stuttgart.de/pub/ug/ug3.1
* email:          ug@ica3.uni-stuttgart.de

    Editor's Note: I put a copy in mgnet/Codes/ug (version 3.1.2) at both Yale
    -------------  and CERFACS.

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

Date: Fri, 10 Nov 1995 09:04:21 -0700
From: cliu@carbon.cudenver.edu (Chaoqun Liu)
Subject: CFD short course

                ADVANCES IN  COMPUTATIONAL FLUID DYNAMICS
                    A 5-Day Short Course Announcement
                Louisiana Tech University, Ruston, Louisiana
                         June 24 - 28, 1996

     The Louisiana Tech University will host a 5-day CFD short course
in the summer of 1996. A number of leading researchers will give lectures
on recent advances in CFD. Several advanced teaching and research codes 
for incompressible flow, direct numerical simulation for flow transition,
and numerical combustion with detailed chemistry will be described and
distributed to all attendants for free.

     Anyone who is interested in CFD and has basic multigrid and CFD
knowledge, including faculty, graduate students, scientific researchers,
government agents, and industrial scientists and engineers,
are welcome to attend.  The registration fee is at a
discounted level of $350 (group rate is $250 and student rate is $150),  
and it covers all  lecture notes, floppy disks  with computational codes, 
a reception, and a formal dinner. 

    The lectures include:

    Multigrid and multilevel adaptive methods 
    First-order system least square (FOSLS)
    Preconditioning for low speed flow
    Central difference, R-K scheme, and residual smoothing
    Boundary conditions for time dependent flows
    Upwind differencing - principles and recent development 
    Advanced turbulence models
    Fast Navier-Stokes solver
    High-order grid generation 
    Direct numerical simulation for complex geometry
    Non-reflecting outflow boundary condition 
    Multilevel grid dissipation
    Numerical simulation for turbulent combustion with detailed chemistry
    Numerical pollutant prediction


    The speakers include:

    Dr. L. Sakell (US Air Force Office of Scientific Research)
    Dr. A. Brandt (Weizmann Institute, Israel)
    Dr. E. Turkel (Tel Aviv University, Israel)
    Dr. P. Roe    (University of Michigan)
    Dr. S. McCormick (University of Colorado at Boulder)
    Drs. T-H Shih and J. Zhu (NASA Lewis Research Center)
    Dr. C-H Sung   (Navy David Taylor Research Center)
    Dr. C. Liu ( Louisiana Tech University)
    Drs. Z. Liu, X. Zheng, and C. Liao ( Louisiana Tech University)


     If you are interested in this short course, please fill
out the attached form and email it back to cliu@math.latech.edu or
send it by regular post to:

     Prof. Chaoqun Liu
     ACFD Short Course Chair 
     Department of Mathematics and Statistics
     Louisiana Tech University 
     P.O. Box 3189, Ruston, 
     LA 71272-0001
     Tel : (318) 257-2538
     Fax : (318)257-3935
     email : cliu@math.latech.edu
     http://www.math.latech.edu/~cliu


For more information please go to:
http://www.math.latech.edu/~cliu
or call Dr. Chaoqun Liu at (318) 257-2924.




________________________________________________________________________

     Name _______________

     Mailing Address :

     _____________________________________________


     _____________________________________________


     _____________________________________________


     Day Phone Number : __________________________

     Fax Number:        __________________________

     Email Address :    __________________________


     I am interested in the ACFD short course. Please send me the

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

Date: Tue, 21 Nov 95 16:05:53 +0300 (MSK)
From: dknv@cmc.msk.su (dknv)
Subject: Book by D'yakonov

Dear Craig, I am pleased to inform you that CRC Press published my book "
Optimization in Solving Elliptic Problems" (about 600 pages) a few weeks ago.
They did an excellent job and it seems that the book is not superexpansive
(about 80 $).  It is possible to contact the Orders department directly by
e-mail (orders@crcpress.com) .  I am asking you to pass the information to
other interested people at your "vicinity".

Best wishes. Gene D'yakonov.

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

Date: Thu, 2 Nov 1995 18:02:04 -0700
From: aknyazev@tiger.cudenver.edu (Andrew Knyazev)
Subject: Faculty opening at CU-Denver

University of Colorado at Denver
Department of Mathematics 

The Department  of Mathematics at the University of Colorado
at Denver  is seeking to fill two tenure-track  positions at 
the  rank of  Assistant Professor,  beginning  August  1996. 
Positions are  contingent on  budgetary approval.   We offer 
B.S.,  M.S.  and  Ph.D. degrees  with  optional focus in one 
of   six   areas:    computational   mathematics,   discrete
mathematics, engineering  mathematics, operations  research,
probability, and  statistics.   These also  comprise  active
areas of research of our resident faculty.

Applicants  are   expected  to  show  strong  potential  for
research and  teaching and have a commitment to professional
service.  Applicants must have, or expect to receive by next
August, a  Ph.D. in mathematics or a related discipline.  In
general, we  seek qualified  applicants in  one of the focus
areas and  prefer someone who can interface with two or more
areas.  For these positions,  the areas of greatest interest 
are  applied  probability or statistics,  and  computational 
mathematics.  However, we  shall consider  applications with 
another focus in applied mathematics.

To apply,  please send a current vita, list of publications,
a statement  of research  plans and  goals, a  statement  of
teaching philosophy,  and arrange  to have  three letters of
recommendation sent to:

             Burt Simon, Search Committee Chair
             Mathematics Department - Campus Box 170
             University of Colorado at Denver
             P.O. Box 173364
             Denver, CO 80217-3364

Our fax  number is  303-556-8550.    For more information on 
our  department  and  university   visit  our  web  site  at
http://www-math.cudenver.edu.     We  will  begin  screening 
applications December  15, 1995.  

The  University  of  Colorado  at  Denver  is  committed  to
enhancing the  diversity of  its administration, faculty and
staff, and  to program access for persons with disabilities,
and invites  and  strongly  encourages  nominations  of  and
applications  from  women,  members  of  ethnic  and  racial
minorities,  veterans,   and   people   with   disabilities.
Alternative formats  of  this  ad  or  job  description  are
available upon request.

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

Date:   Mon, 20 Nov 1995 09:39:49 -0500
From: Christopher Beattie 
Subject: Faculty opening at Virginia Tech

We posted this on nanet a few weeks ago but want to make sure we have
healthy representation of multigrid types in our applicant pool.

   Virginia Polytechnic Institute and State University
   
                 Department of Mathematics
   
   Applications are solicited for a tenure-track assistant
  professorship in numerical analysis. Candidates must have
  a PhD in mathematics or equivalent, with a strong record
  or demonstrated potential in research and teaching. We seek
  candidates who will augment an existing faculty focus in
  numerical treatment of PDEs, optimization, and computational
  linear algebra.
   
   Send a letter of application, curriculum vitae, summary
  of research plans, together with four letters of recommendation
  (one of which addresses teaching skills), to:
        Numerical Analysis Search Committee,
        Department of Mathematics,
        Virginia Tech,
        Blacksburg, VA 24061-0123. 
  Review of applications will begin on December 20, 1995
  and will continue until the position is filled.
  
   Virginia Tech has a strong commitment to the principle of diversity
  and, in that spirit, seeks a broad spectrum of candidates including
  women, minorities, and people with disabilities.  Individuals with
  disabilities desiring accommodations in the application process should
  contact Christopher Beattie, Department of Mathematics, 540-231-6536
   (TDD/PC 1-800-828 1120- Voice 1-800-828-1140).

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

Date: Mon, 27 Nov 1995 08:08:08 -0500
From: Craig Douglas 
Subject: Updated preprint by Douglas

               A Review of Numerous Parallel Multigrid Methods

                               Craig C. Douglas

                       IBM T. J. Watson Research Center
                 Yale University Computer Science Department
                                   CERFACS

                                   Abstract

Multigrid methods originated earlier this century, in the personnel computing
era.  Someone who needed to compute an approximation to the solution of a
partial differential equation during that era would fill a room with people.
After using very simple mechanical calculators to compute parts of the
approximation, these people would pass their parts to the other people in the
room who needed them.  Except for the very different time scales and
approximate solution accuracy, this process is similar to computing on today's
distributed memory parallel computers.

Parallel multilevel methods are shown to be the natural precursors to standard
multilevel methods based on the personnel computing era of earlier this
century.  They are also the natural successors to standard multilevel methods
in the age of computers.  What makes six parallel multilevel methods practical
and impractical is discussed in the context of the three algorithms that
encapsulate them.

Appears in slightly edited form in SIAM News, V.25, N.3, May, 1992.
This will appear in

    Applications on Advanced Architecture Computers
    Edited by Greg Astfalk
    Available spring 1996
    Approx. 350 pages
    Softcover
    ISBN 0-89871-368-4
    List Price $35.00
    SIAM Member Price $28.00
    Order Code OT53

    Editor's Note: in mgnet/papers/Douglas/para-mg.ps.gz and
    -------------     mgnet/papers/Douglas/para-mg.abs.

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

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