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

- Matrix Algorithms, Volume 2
- Computing Volumes from Areas
- PseudoPack, A Software Library for Numerical Differentiation
- SIAM Conference on Applied Linear Algebra
- Bay Area Scientific Computing Day
- Contents, Constructive Approximation

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

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

From: G. W. Stewart <stewart@cs.umd.edu>

Date: Mon, 27 Dec 1999 14:06:25 -0500 (EST)

**Subject: Matrix Algorithms, Volume 2**

I am about half through the second volume of my series Matrix

Algorithms. It is titled Eigensystems and concerns the treatment of

dense and sparse eigenvalue problems. I have just posted the first

three chapters--Eigensystems, The QR Algorithm, and The Symmetric

Eigenvalue Problem--along with front material, an appendix, and a

bibliography. They may be obtained though my home page at

http://www.cs.umd.edu/~stewart/

or at

ftp://thales.cs.umd.edu/pub/survey/

They may also be obtained by anonymous ftp at thales.cs.umd.edu in

pub/survey.

If you have comments, suggestions, or errata please send them

to me at

stewart@cs.umd.edu

G. W. (Pete) Stewart

Department of Computer Science

University of Maryland

College Park, MD 20002

USA

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

From: Joel Malard <JM.Malard@pnl.gov>

Date: Wed, 29 Dec 1999 12:00:42 -0800

**Subject: Computing Volumes from Areas**

Dear Na-Neters,

The area A of a triangle can be found from the

lengths x1, x2 and x3 of its sides using the formula:

A = 0.25*\sqrt{ 4*x1^2*x2^2 + 4*x1^2*x3^2 +

4*x2^2*x3^2 - (x1^2+x2^2+x3^2)^2 }.

Is there a more robust formula? In the end i would like

to compute the volume of a tetrahedron from the areas

of its facets? I thank you for your help.

Best Wishes,

Joel Malard

Pacific Northwest National Laboratory

Battelle Boulevard, PO Box 999

Richland, WA 99352

USA

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

From: Wai Sun Don <wsdon@cfm.brown.edu>

Date: Sat, 1 Jan 2000 17:50:46 -0500 (EST)

**Subject: PseudoPack, A Software Library for Numerical Differentiation**

Hello, this message announces release of the PseudoPack 2000 Rio Edition.

PseudoPack is a software library for numerical differentiation by

pseudospectral methods.

It is being developed by Prof. Wai Sun Don at Brown University and

Prof. Bruno Costa at Departamento de Matematica Aplicada, IM-UFRJ, Brazil.

This special edition of the PseudoPack is dedicate to honor our dear friend and

colleague Prof. David Gottlieb. His pioneer work on Pseudospectral methods with

Prof. Steve Orzag in late 1970 leads to greater appreciation of the value of

high order scheme within the scientific computing community.

Without his guidance and encouragement, PseudoPack would be a Pseudo-reality.

Brief description of features :

1. Derivatives of up to order four are supported for the Fourier, Chebyshev

and Legendre collocation methods that are based on the Gauss-Lobatto,

Gauss-Radau and Gauss quadrature nodes.

2. Matrix-Matrix Multiply, Even-Odd Decomposition and Fast Fourier Transform

Algorithms are supported for computing the derivative/smoothing of

a function.

3. Native fast assembly library calls such as General Matrix-Matrix Multiply

(GEMM) from Basic Linear Algebra Level 3 Subroutine (BLAS 3),

Fast Fourier Transform (FFT) and Fast Cosine/Sine Transform (CFT/SFT)

when available, are deployed in the computational kernel of the PseudoPack.

4. Special fast algorithms, e.g. Fast Quarter-Wave Transform and Even-Odd

Decomposition Algorithm, are provided for cases when the function

has either even or odd symmetry.

5. Kosloff-Tal-Ezer mapping is used to reduce the roundoff error

for the Chebyshev and Legendre differentiation.

6. Extensive built-in and User-definable grid mapping function suitable

for finite, semi-infinite and infinite domain are provided.

7. Built-in filtering (smoothing) of a function and its derivative

are incorporated in the library.

8. Differentiation and Smoothing can be applied to either the first

or the second dimension of a two-dimensional data array.

9. Conservative and non-Conservative form of Derivative operators, namely,

Gradient, Divergence, Curl and Laplacian operators in the 2D/3D general

curvilinear coordination using pseudospectral methods are available.

10. Unified subroutine call interface allows modification of any aspect

of the library with minor or no change to the subroutine call statement.

It aims are to provide minimum roundoff error and good efficiency on

several computational platforms. They are SGI, Sun, IBM R6000 and Cray.

The software package is written in Fortran 90 with the C preprocessor.

It is freely available, at least to non-commercial users.

You can read the full description of the library from our Web page

http://www.labma.ufrj.br/~bcosta/PseudoPack2000/Main.html

or

http://www.cfm.brown.edu/people/wsdon/home.html

To obtain the package, please contact Prof. Bruno Costa at

bcosta@labma.ufrj.br

In the month of January 2000, please contact Prof. Wai Sun Don at

wsdon@cfm.brown.edu

instead since Prof. Bruno Costa will be on summer vacation.

Please send any questions, comments, or suggestions to wsdon@cfm.brown.edu or

bcosta@labma.ufrj.br

Thanks for your time.

-- Prof. Bruno Costa and Prof. Wai Sun Don

January 1, 2000

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

From: Darrell Ross <ross@siam.org>

Date: Mon, 27 Dec 1999 10:17:03 -0500

**Subject: SIAM Conference on Applied Linear Algebra**

Greetings,

I'm writing to inform the NA Digest that the Seventh SIAM Conference

on Applied Linear Algebra (LA00) call for papers is

now on the web at:

http://www.siam.org/meetings/la00/

The deadline for submission of contributed abstracts for a poster

presentation or lecture presentation is May 1, 2000.

Electronic submission are welcome using the new Conference Management

System at:

http://www.siam.org/meetings/la00/part.htm

Please feel free to contact me if you have any questions.

Regards,

Darrell Ross, Conference Program Manager

Society for Industrial & Applied Mathematics

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

From: Esmond Ng <EGNg@lbl.gov>

Date: Thu, 30 Dec 1999 21:43:20 -0800

**Subject: Bay Area Scientific Computing Day**

BAY AREA SCIENTIFIC COMPUTING DAY

Saturday, February 26, 2000

8:45 am - 5:15 pm

Lawrence Berkeley National Laboratory

Building 66

One Cyclotron Road

Berkeley, CA 94720

GOAL:

The Bay Area Scientific Computing Day will be an informal gathering to

encourage the interaction and collaboration of researchers in the field

of scientific computing from the Bay Area. It is also meant to provide

students and young researchers an opportunity to gain experience in

presenting their work to a large group setting.

FORMAT:

The one-day workshop will begin at 8:45am with some opening remarks.

Technical presentations will begin at 9:00am. The workshop will finish at

about 5:15pm. All presentations will be 30 minutes long. This will allow

us to accommodate a total of 13 talks.

SPEAKERS:

Speakers will be selected from the three major DOE labs in the Bay Area

(Berkeley Lab, Lawrence Livermore Nat'l Lab, and Sandia Nat'l Lab),

Stanford Linear Accelerator Center, as well as from Stanford University

and University of California at Berkeley.

SPECIAL POST-WORKSHOP EVENT:

Tentative evening banquet at a local restaurant (location and time to be

announced).

REGISTRATION:

The workshop registration form is available on WWW

(http://www.nersc.gov/research/SCG/forms2.html).

It is also available from Eric Essman (epessman@lbl.gov, fax: 510-486-5812).

If you are interested in attending the workshop, please email or fax the

registration form to Eric Essman no later than January 19, 2000.

ADDITIONAL INFORMATION:

Additional information is available on WWW

(http://www.nersc.gov/research/SCG/bascd.html).

Also, please contact Gene Golub (golub@sccm.stanford.edu) or Esmond G. Ng

(egng@lbl.gov; Phone: 510-495-2851) if you have further questions.

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

From: Kristen Dranikoski <ca@math.usf.edu>

Date: Wed, 29 Dec 1999 15:52:09 -0500 (EST)

**Subject: Contents, Constructive Approximation**

CONSTRUCTIVE APPROXIMATION CONTENTS

Vol. 16, Number 2, 2000

161-199 Y.A. Brudnyi and N.J. Kalton

Polynomial Approximation on Convex Subsets of R^n

201-219 G.C. Donovan, J.S. Geronimo, and D.P. Hardin

Compactly Supported, Piecewise Affine Scaling Functions

on Triangulations

221-259 W. Dahmen, B. Han, R.-R. Jia, and A. Kunoth

Biorthogonal Multiwavelets on the Interval: Cubic Hermite

Splines

261-281 Vilmos Totik

Weighted Polynomial Approximation for Convex External Fields

283-311 P.M. Soardi

Biorthogonal M-Channel Compactly Supported Wavelets

RESEARCH PROBLEMS

313-316 D.S. Lubinsky

Asymptotic Behaviour of Entire Functions with Positive

Coefficients: Research Problems 2000-2

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

End of NA Digest

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

-------