Memorandum

To: Caroline Wardle
From Lloyd Fosdick
Subject: Report of EI activities and dissemination of materials
Date: 14 December 1992

	This memorandum is in response to your e-mail request of
18 November.

	The text material we have produced for the HPSC course is in
two major parts.  The first part includes tutorials and other
reference material as listed below:

1. An Overview of Scientific Computing (A)

2. The Cray Y-MP: An Introduction (B)

3. The Connection Machine CM-2: An Introduction (A)

4. The Intel iPSC/2: An Introduction (C)

5. Molecular Dynamics: A Tutorial (A)

6. Tomography: A Tutorial (B)

7. Advection: A Tutorial  (C)

8. Performance Measurement: A Tutorial (C)

9. Scientific Visualization: A Tutorial (C)

10. Elements of AVS (A)

11. Elements of IDL (A)

12. Elements of Matlab (A)

13. Fortran Short Reference (B)

14. vi Short Reference (A)

15. Unix Short Reference (A)

16. IEEE Arithmetic Short Reference (A)

17. A Review of Selected Topics from Numerical Analysis (A)


The second part consists of a Laboratory Manual with
twenty-three chapters as listed below:

1. Matlab (A)

2. Visualization (A)

3. Measuring Operation Time (A)

4. Cray Y-MP Performance (B)

5. Intel iPSC/2 Performance (B)

6. Connection Machine CM-2 Performance (A)

7. Application Visualization System (AVS) (A)

8. Molecular Dynamics: Chain of Atoms (B)

9. Molecular Dynamics: Accuracy of Numerical Methods (B)

10. Molecular Dynamics on the Cray Y-MP (B)

11. Molecular Dynamics on the iPSC/2 (B)

12. Molecular Dynamics on the CM-2 (B)

13. Advection: Explicit Finite Difference Algorithm (B)

14. Advection: Implicit Finite Difference Algorithm (B)

15. Advection on the Cray Y-MP (B)

16. Advection on the iPSC/2 (C)

17. Advection on the CM-2 (C)

18. Particle Advection (C)

19. Tomography: Geometrical Considerations (B)

20. Tomography: Fast Fourier Transform (C)

21. Tomography: Image Reconstruction (C)

22. Tomography on the iPSC/2: Fast Fourier Transform (C)

23. Tomography on the iPSC/2: Image Reconstruction (C)


The letters A, B, or C at the end of each title indicate the state of
completion of the work: A indicates that the material is complete and
is being distributed; B indicates that the item is in production; C
indicates that the material is scheduled for production. Most items
now listed as B will move to status A by the middle of February 1993;
most items now listed as C will move to status A by the middle of
summer 1993.  Preliminary notes of some items marked B and C have been
used in class and workshops.

In addition to the text material listed above we have developed a
large amount of software. Each chapter of the laboratory manual
contains a number of examples and exercises for which software has
been developed. This includes software written for the workstations,
the Cray Y-MP, the CM2, and the iPSC/2. Most of the software is
written in Fortran; some is in the form of IDL and Matlab scripts; and
some is in the form of modules for AVS, written in C.

We have disseminated our materials through workshops, lectures at
conferences and other schools, U. S. Post, and anonymous ftp.  The
anonymous ftp address is "cs.colorado.edu", and our completed material
is in the directory "pub/cs/HPSC". Also, because our project involves
collaboration with five schools (U. New Mexico, U. Texas at San Antonio,
New Mexico Tech, Fort Lewis College, and Colorado School of Mines) we
have had many interactions with our collaborators at those
schools through meetings, phone conversations, e-mail,
and so forth. The software has only been made available to our collaborators.
We don't expect to distribute it generally until mid summer 1993.

Formal talks have been given at:
	1. International Federation of Information Processing Societies,
	Working Group 2.5 Toronto Meeting (Sept. 1992).
	2. Supercomputing '92, Minneapolis (Nov. 1992).
	3. Mesa State College (Oct. 1992).
	4. Colorado School of Mines (Nov. 1992).

And we have a paper scheduled at:
	5. SIAM Conference on Parallel Processing for Scientific Computing,
	Norfolk (Mar. 1993).

Three workshops have been held:
	1. Summer 1991: A two-week workshop for our collaborators at
	U. New Mexico, U. Texas at San Antonio, New Mexico Tech, 
	Fort Lewis College, and Colorado School of Mines. 

	2. Summer 1992: A two-week workshop for undergraduates. There 
	were fifteen students from the following schools: U. New Hampshire,
	U. Vermont, U. Idaho, Colorado School of Mines, Colorado State U.,
	U. Texas at San Antonio, New Mexico Tech., U. New Mexico, U. Wisconsin.

	3. Summer 1992: A two-week workshop for faculty. There were nine
	faculty members from the following schools: Mesa State College,
	U. New Hampshire, U. Vermont, Drexel U., U. Alabama at Birmingham,
	U. Nevada at Las Vegas, San Diego State U., New Mexico Tech., 
	U. Texas at San Antonio. Another faculty member (from U. New Mexico)
	attended the student workshop.

Based on the exit evaluations we received from these workshops they
were very successful. Two two-week workshops for students and faculty
are scheduled for Summer 1993. They will be similar to the workshops
held in summer 1992.

I don't think we have had enough time to determine which method of
dissemination is most successful. We only started our ftp distribution
a couple of months ago. As of now, our feeling is that personal
contact through workshops and meetings is an essential part of the
distribution process. Certainly the early collaboration we established
with five schools gave us an excellent built-in mechanism for
distribution.  As a direct result of the workshops from last summer a
one-semester course based on our materials will be offered at U.
Nevada Las Vegas; and it seems likely that Drexel U. will offer a
course using some of our materials next year. At the collaborating
schools there has been more activity. Our materials were used at Fort
Lewis College in a course in Spring 1992. This Fall our materials are
being used at U. New Mexico, and New Mexico Tech. U. Texas at San
Antonio plans to use our materials in Spring 1993.

One important thing I should mention concerning the issue of
distribution is that we had to spend much more time in giving support
to the collaborating institutions than we anticipated.  We provided
each collaborator with a DEC 5000 workstation, an X-terminal, an
Exabyte tape drive and some licensed software (Matlab, Mathematica,
NAG Library), and software we developed. Most of them did not have
adequate technical support on site for installation of equipment and
software.  We tried to help them as much as possible through telephone
conversations and e-mail. These efforts plus visits they made to our
site were successful, but the time spent on this was much more than we
anticipated. Visits by persons from our technical staff to the
collaborating sites would probably have been very helpful but we did
not have the resources to provide this level of support.

In addition to the HPSC course we are also developing a course in
Scientific Visualization. Information on this course was used at a
workshop organized by G. Domik (a co-PI) at the IEEE Visualization '92
Conference in Boston, October 1992. The course content is described in
detail in a paper submitted to IEEE Transactions on Education.
Information on this course is available by anonymous ftp to
cs.colorado.edu and is located in pub/cd/vis .