This includes many, but certainly not all, of the key CP participants. The bibliography and Appendix A cites the full set of CP reports and authors.

**Giovanni Aloisio**

Dipartimento di Elettrotecnica ed Elettronica

Facolta di Ingegneria-Politecnico di Bari (Italy)

Via Orabona, 4

70125 Bari (Italy)

Aloisio@vaxle.le.infn.it

Worked from (11/86-end of project):
Investigating the efficiency of the Hypercube architecture in Real-Time
SAR data processing (``SAR Hypercube Project''). Non traditional FFT
algorithms, such as the Prime Factor, have been coded to run on the
nCUBE, iPSC, and Mark IIIfp hypercubes. The optimal decomposition, on
a specific hypercube system, of a complete software package for digital
SAR data processing has been determined. This package has been
implemented in the sequential version on a VAX-780 at IESI/CNR
(Bari-Italy) and has been tested on digital raw data obtained by JPL
(SIR-B space Shuttle mission).
Now works on:
High Performance Distributed Computing (porting of several
applications under PVM and Net-EXPRESS. Parallel compilers, such as
HPF, will also be tested). A joint project with CCSF is in progress.

**Ian Angus**

Research Scientist

Boeing Computer Services

P. O. Box 24346, MS 7L-48

Seattle, WA 98124-0346

angus@atc.boeing.com

Worked from (1986-1987):
Involved primarily with the implementation of a Hypercube simulator
and with the design and first implementation of the Fortran Cubix
programming system.
Now works on:
Programming tools and environments, object oriented approaches to scientific
and parallel computing, and compilation of object oriented languages.

**John Apostolakis**

CERN

CN Division, 513-R-024

CH 1211 GENEVA 23, Switzerland

japost@dxcern.cern.ch

Worked from (9/86-end of project):
With lattice gauge theory, lattice spin models, and gravitational
lenses and the issues involved in developing efficient parallel
programs to simulate them.
Now works on:
Implementing experimental high energy physics applications on Massively
Parallel Processors.
Contributed Section 7.4, Statistical Gravitational
Lensing

**Clive F. Baillie**

Research Fellow

Computer Science Department

Campus Box 430

University of Colorado

Boulder CO 80309

clive@kilt.cs.colorado.edu

Worked from (9/86-end of project):
Implementations of physics problems, particularly clustering methods and
performance studies. Large-scale Monte-Carlo simulations of QCD, XY and
O(3) models, 3D Ising model, 2D Potts model and dynamically triangulated
random surfaces (DTRS).
Now works on:
Further work on DTRS, making them self-avoiding to simulate superstrings,
and adding Potts models to simulate quantum gravity coupled to matter.
Contributed Sections 4.3, Quantum Chromodynamics;
4.4, Spin Models; 7.2, Dynamically Triangulated Random Surfaces; and
12.6, Cluster Algorithms for Spin Models

**Vasanth Bala**

Member of the Technical Staff

Kendall Square Research

170 Tracer Lane

Waltham, Massachusetts 02154

vas@ksr.com

Worked from (8/89-end of project):
With the design of software tools,
compiler optimizations, and communication libraries for scalable parallel
computers.
Now works on:
Speculative instruction scheduling for superscalar RISC processors, and
general compiler optimization of C, Fortran90/HPF and C++ programs for
RISC-based parallel computers. After leaving Caltech CP, was a research
staff member at IBM T. J. Watson Research Center (Yorktown Heights, NY)
involved in the design of the IBM SP1 parallel computer.
Contributed Section 13.2, A Software Tool

**Ted Barnes**

Staff Physicist

Theoretical Physics Division

Oak Ridge National Laboratory

Oak Ridge, Tennessee 37831-8083

and
Associate Professor of Physics

Department of Physics

University of Tennessee

Knoxville, Tennessee 37996

Worked from (1987-1989):
Monte Carlo calculations to simulate high-temperature superconductivity.
Now works on:
QCD spectroscopy, couplings and decays of hadrons, high-temperature
superconductivity.
Contributed Section 7.3, Numerical Study of
High- Spin Systems

**Roberto Battiti**

Assistant Professor of Physics

Universita` di Trento

Dipartimento di Matematica

38050 Povo (Trento), Italy

battiti@itnvax.cineca.it

Worked from (1986-end of project):
Parallel implementation of neural nets and vision algorithms; computational
complexity of learning algorithms.
Now works on:
Constructive and destructive learning methods for neural nets, ``natural''
problem solving such as genetic algorithms; application of neural nets
in financial and industrial areas.
Contributed Sections 6.5, A Hierarchical Scheme for
Surface Reconstruction and Discontinuity Detection; 6.7, An Adaptive
Multiscale Scheme for Real-Time Motion Field Estimation; 6.8,
Collective Stereopsis, and 9.9, Optimization Methods for Neural Nets:
Automatic Parameter Tuning and Faster Convergence

**Jim Bower**

Associate Professor of Biology

Computation and Neural Systems Program

California Institute of Technology

Mail Code 216-76

Pasadena, California 91125

jbower@smaug.bbb.caltech.edu

Worked from (1988-end of project):
Using concurrent computers to build large-scale realistic
models of the nervous system. We recognized early on that
truely realistic models of these complex systems would
require the power present in parallel computation. This, in
fact, is reflected in the fact that the nervous system
itself is probably a parallel device. Leader of GENESIS project
described in Section 7.6.
Now works on:
Current interest remains understanding the relationships
between the structure and the function of the nervous
system. We have recently published several scientific
papers that would have not been possible without the use of
the concurrent machines at Caltech.

**Eugene D. Brooks, III**

Deputy Associate Director

Advanced Technologies Computation Organization

Lawrence Livermore National Laboratory

P. O. Box 808, L-66

Livermore, CA 94550

brooks3@llnl.gov

Worked from (1981-1983):
The use of parallel computing to supply a new computational
capability for computational physics tasks.
Now works on:
Parallel computer architecture, parallel languages, computational
physics algorithms, and parallelization of computational physics
algorithms.

**Robert W. Clayton**

Professor of Geophysics

California Institute of Technology

Geophysics, 350 S. Mudd

Mail Code 252-21

Pasadena, CA 91125

clay@seismo.gps.caltech.edu

Worked from (1983-end of project):
Finite-difference solutions of wave phenomena. Imaging with
seismic reflection data.
Now works on:
Finite-difference solutions of wave phenomena. Imaging with
seismic reflection data.
Contributed Section 18.2, ISIS: An Interactive
Seismic Imaging System

**Paul Coddington**

Syracuse University

Northeast Parallel Architectures Center

111 College Place, 3-228 CST

Syracuse, New York 13244-4100

paulc@npac.syr.edu

Worked from (1988-end of project):
Developed parallel implementations of non-local Monte Carlo algorithms for
spin models of magnetism.
Now works on:
From 1990-92, worked as a Research Associate at NPAC on computational
physics applications, including new sequential and parallel Monte Carlo
algorithms for spin models and dynamically triangulated random surface
models of quantum gravity, as well as parallel algorithms for connected
component labeling and graph coloring. Also worked on improved
stochastic optimization techniques, such as simulated annealing.

From 1992 until the present, worked as a Research Scientist at NPAC leading a project on the use of parallel computing in the power utility industry. This involves porting existing code to parallel computers, and developing parallel algorithms for sparse matrix computations and differential-algebraic equation solvers.

**Dave Curkendall**

ALPHA Project Manager and

Advanced Parallel Processing Program Manager

Jet Propulsion Laboratory

4800 Oak Grove Drive, MC 138-310

Pasadena, California 91109

DAVE_CURKENDALL@macq_smtp.Jpl.Nasa.Gov

Worked from (8/84-end of project):
As Hypercube Task Manager and later as Hypercube Project Manager, was
interested in the hypercube hardware development, its operating system,
particularly the asynchronous message-passing developments of Mercury
and Centaur, and in the development of large-scale simulations.
Now works on:
The development of discrete event simulation software for parallel
machines and techniques for the remote, interactive exploration of
large, image and geographical databases.

Contributed Section 18.3, Parallel Simulations that Emulate Function

**Hong-Qiang Ding**

Member of Technical Staff

Jet Propulsion Laboratory

4800 Oak Grove Drive

Mail Stop 169-315

Pasadena, California 91109

hding@redwood.jpl.nasa.gov

Worked from (8/87-end of project):
Extensive and large-scale simulations QCD and quantum spin models.
Now works on:
Developing efficient methods for long-range interactions and molecular
simulations; simulate model superconductors with parallel machines.
Contributed Sections 6.3, Magnetism in the
High-Temperature Superconductor Materials; and 6.4, Phase Transitions
in Two-dimensional Quantum Spin Systems

**David Edelsohn**

IBM T. J. Watson Research Center

P. O. Box 218

Yorktown Heights, NY 10598-0218

c1dje@watson.ibm.com

Worked from (1989-end of project):
Computational astrophysics simulations of galaxy formation and
evolution, and cosmology using concurrent, multiscale, hierarchical
N-body and adaptive mesh refinement algorithms.
Now works on:
As a doctoral candidate at the Northeast Parallel Architectures Center,
Syracuse University, his research interests include computational
astrophysics simulations of galaxy formation and evolution, and
cosmology using concurrent, multiscale, hierarchical N-body and adaptive
mesh refinement algorithms; and object-oriented concurrent languages.
He is visiting IBM as an IBM Computational Science Graduate Fellow.
Contributed Section 12.8, Hierarchical
Tree-Structures as Adaptive Meshes

**Ed Felten**

Assistant Professor

Department of Computer Science

Princeton University

35 Olden Street

Princeton, New Jersey 08544

felten@cs.princeton.edu

Worked from (1984-end of project):
Research interests included a variety of issues surrounding how to
implement irregular and non-numerical applications on
distributed-memory systems.
Now works on:
How to build system software for parallel machines, and how to
construct parallel programs to use this system software. More
generally, my research interests include parallel and distributed
computing, operating systems, architecture, and performance modeling.

**Jon Flower**

President

ParaSoft Corporation

2500 E. Foothill Blvd.

Pasadena CA 91107

jwf@parasoft.com

Worked from (1983-end of project):
High-energy physics simulations; programming tools, debugging and
visualization. Founder and President of ParaSoft Corporation
Now works on:
Programming environments, tools, libraries for parallel computers.
Contributed Sections 5.2, A ``Packet'' History of
Message-passing Systems; 5.3, Parallel Debugging; 5.4, Parallel
Profiling; and 13.5, ASPAR

**Geoffrey C. Fox**

Professor of Computer Science and Physics

Director, Northeast Parallel Architectures Center

Syracuse University

111 College Place

Syracuse, New York 13244-4100

gcf@npac.syr.edu

Worked from (1981-end of project):
Involved as Principal Investigator with particular attention to
applications, algorithms, and software. Developed the theory of problem
architecture to describe and classify results of CP. Developed concepts
in computational science education based on student involvement in CP
and implemented new curricula initially at Caltech and later at Syracuse
University.
Now works on:
From 1990 until the present, directs the project at Syracuse University,
which has a similar spirit to CP, but is aimed more at industry than at
academic problems.
Contributed Chapters 1, 3, 19, and 20; Sections 4.1,
4.2, 5.1, 6.1, 7.1, 9.1, 11.2, 11.3, 12.1, 13.1, 13.3, 13.7, 14.1, 15.1, and
18.1

**Sandy Frey**

President, Reliable Distributed Information Corporation

Pasadena, CA 91107

sandy@ccsf.caltech.edu

Worked from (1984-1988):
Studying the system problems of implementing a teraflop machine
with 1980s technology, and the data management problems involved
in implementing massive data intensive applications in parallel
processing environments, such as hypercubes.
Now works on:
Data management problems involved in implementing massive data
intensive applications in parallel processing environments, such
as hypercubes.

**Wojtek Furmanski**

Research Professor of Physics

Syracuse University

201 Physics Building

Syracuse, New York 13244-1130

furm@npac.syr.edu

Worked from (1986-end of project):
Developed a class of optimal collective communication algorithms
implemented on Caltech hypercubes, and applied in parallel implementation of
neuroscience simulations and machine vision algorithms.
Now works on:
Based on lessons learned in these early parallel simulations, developed
MOVIE system aimed at a general purpose platform for interactive HPCC
environments. MOVIE, initially used for terrain image analysis, is now
further developed at NPAC. Recently, the HPF interpreter has been
constructed on top of MOVIE, and the MOVIE system is now further developed
with the aim of integrating HPCC and Virtual Reality software technologies
towards the broadband network based televirtuality environment.

Contributed Chapter 17, MOVIE - Multitasking
Object-oriented Visual Interactive Environment
**Jeff Goldsmith**

California Institute of Technology

Mail code 350-74

Pasadena, California 91125

jeff@gg.caltech.edu

Worked from (1985-end of project):
Computer Graphics.
Now works on:
Computer Graphics, in particular, computer-designed motion.

**Peter Gorham**

Project Manager

University of Hawaii at Manoa

Honolulu, Hawaii 96822

gorham@fermion.phys.Hawaii.Edu

Worked from (1987-end of project):
My work with P came about through Tom Prince's involvement
with the project. Tom hired me as a postdoc in 1987 and I arrived in
February of that year. Tom was beginning a collaboration with
Shri Kulkarni of the Caltech Astronomy Department in two areas:
first, a program to develop code for bispectral anlaysis of
astronomical speckle interferograms taken with the Hale
telescope; and second, a search for new radio pulsars using the
Arecibo Observatory's transit telescope. In both cases,
the telescopes involved were among the largest of their class and the
data sets to be produced could only be managed with a supercomputer.
Also in both cases, the data analysis lent itself very well to parallel
processing techniques.

Both programs were very successful and Tom and I had the pleasure of seeing two graduate students complete their PhD requirements in each of the research areas (Stuart Anderson, pulsars; and Andrea Ghez, infrared speckle interferometry). Something of order a dozen research papers came out of this effort before I left for my present position in July of 1991, and a steady stream of results have come out since. Now works on: The Deep Underwater Muon and Neutrino Detector (DUMAND) project. This project is developing a large, deep ocean Cherenkov detector which will be sensitive to high energy neutrino interactions and will have the capability to produce images of the sky in the ``light'' of neutrinos, with angular resolution of order 1 degree. The motivation behind such research arises from current belief that emission of high energy neutrinos may be a dominant process by which active galactic nuclei and QSOs release energy into their galactic environment. Detection of such neutrinos would provide unique information about the central engine of such galaxies.

**Thomas D. Gottschalk**

Member of the Professional Staff

California Institute of Technology

Mail code 356-48

Pasadena, California 91125

tdg@cithex.cithep.caltech.edu

tdg@bigbird.jpl.nasa.gov

Worked from (1987-end of project):
Concurrent multi-target tracking for SDI scenarios/applications.
Now works on:
Multi-target tracking (aircraft and space objects), surveillance systems
operations, including sensor tasking, and design rule checking for VLSI
systems.
Contributed Sections 9.8, Munkres Algorithm for
Assignment; and 18.4, Multi-Target Tracking

**Gary Gutt**

Member of the Technical Staff

Jet Propulsion Laboratory

4800 Oak Grove Drive

Mail Stop 183-401

Pasadena, California 91109

gmg@mg.jpl.nasa.gov

Worked from (4/88-5/89):
Numerical simulation of granular systems using the lattice grain dynamics
paradigm.
Now works on:
Microgravity containerless materials processing; development of electrostatic
and electromagnetic positioning techniques for use in microgravity
containerless materials processing.
Contributed Section 4.5, An Automata Model for
Granular Materials

**Peter Halamek**

Technical Staff Member

Jet Propulsion Laboratory

Mail Stop 301-125L

Pasadena CA 91109

pxh@hamlet.caltech.edu

Worked from (6/88-1/89):
Image processing; determination of 3D physical properties of objects from
2D camera images taken aboard a spacecraft.
Now works on:
Optical navigation related research: improving accuracy of extended body
center-finding on images of celestial bodies.

**Paul G. Hipes**

Vice President

Salmon Brothers, Inc.

7 World Trade Center

37th Floor

New York, New York 10048

hipes@daffy.sbi.com

Worked from (11/87-end of project):
Direct solvers for dense systems of linear equations, special purpose
matrix O.D.E. solvers, electron-molecule scattering problems
approached with Schwinger variational methods, atom-molecule
scattering problems approached by direct expansion methods, and green
function Monte Carlo techniques for stationary states of many-electron
systems.
Now works on:
the term structure of interest rates and related topics in fixed
income arbitrage.

**Alex Ho**

Research scientist

IBM Almaden Research Center

K54/802

650 Harry Rd.

San Jose, California 95120-6099

Worked from (7/85-end of project):
Pattern recognition, artificial intelligence, neural nets, robot navigation.
Now works on:
Massively parallel computing, programming models, architectures,
fault-tolerance, performance evaluation.

**Mark A. Johnson**

Senior Engineer/Scientist

IBM Corporation

Internal Zip 4441

11400 Burnet Road

Austin, Texas 78758

maj@austin.ibm.com

Worked from (1983-1986):
Pursued research that led to a Ph.D. in Statistical Physics. Primary
research interests included studying melting in a two-dimensional system
of interacting particles.
Now works on:
System architecture in the area of High End Technical Systems of the
Advanced Workstations and Systems Division of IBM.
Contributed Section 14.2, Melting in Two Dimensions

**Jai Sam Kim**

Associate Professor, Department of Physics

Pohang Institute of Science and Technology

Hyoja-dong San 31

Pohang 780-784, S. KOREA

jsk@vision.postech.ac.kr

Worked from (1986-1988):
Involved in the development of the hypercube simulator NSIM. Later,
he parallelized the FFT codes with Italian visitors Aloisio and
collaborators. Their work on the prime factor DFT code demonstrated
the usefulness of Crystal_Router and also the limitations with the
store-and-forward routing method. He wrote the FORTRAN application
codes that were included in *Solving Problems on Concurrent
Processors*, Vol. 2 [Angus:90a].
Now works on:
Shortly before he returned to his home country Korea, he joined the
interactive parallelizer project described in Chapter 13. He has not
been heard from for some time, but has recently parallelized some
working PDE codes used by mechanical engineers both on NSIM and PVM.
**Adam Kolawa**

Chairman/CEO ParaSoft Corporation

2500 E. Foothill Blvd., Suite 104

Pasadena, California 91107

ukola@flea.parasoft.com

Worked from (1983-end of project):
Development of system software for parallel computers.
Now works on:
Development of software tools.

**Jeff Koller**

Computer Scientist

Information Sciences Institute

4676 Admiralty Way

Marina del Rey, California 90292

koller@isi.edu

Worked from (1987-1989):
MOOS II operating system, application of novel optimization techniques to
dynamic load balancing and compiler optimization.
Now works on:
VLSI design and system software for next-generation parallel machines.
Contributed Sections 13.4, Optimizing Compilers by
Neural Networks; and 15.2, MOOS II: An Operating System for Dynamic
Load Balancing on the iPSC/1

**Aron Kuppermann**

Professor

California Institute of Technology

Mail Code 127-72

Pasadena, California 91125

aron@caltech.edu

Worked from (from beginning to end of project):
Quantum mechanical reaction dynamics; reactive scattering methodologies
suitable for MIMD machines.
Now works on:
Adapting quantum mechanical reaction dynamics codes to new parallel
machines.
Contributed Section 8.2, Quantum Mechanical
Reactive Scattering using a High Performance Parallel Computer

**Paulette C. Liewer**

Member of the Technical Staff

Jet Propulsion Laboratory

4800 Oak Grove Drive

Mail Stop 198-231

Pasadena, California 91109

pauly@hyper-spaceport.jpl.nasa.gov

and
Visiting Associate in Applied Physics

California Institute of Technology

Mail Code 128-95

Pasadena, California 91125

Worked from (1986-end of project):
Concurrent algorithms for particle-in-cell codes.
Now works on:
3D plasma particle-in-cell codes; application of concurrent PIC codes to
problems in solar, space and laboratory plasmas.
Contributed Section 9.3, Plasma Particle-in-Cell
Simulation of an Electron Beam Plasma Instability

**Gregory A. Lyzenga**

Associate Professor of Physics

Harvey Mudd College

Physics Department

Claremont, California 91711

lyzenga@hmcvax.ac.hmc.edu

Worked from (1985-end of project):
Parallel solution of fine element problems as applied to
geophysics, solid mechanics, fluid mechanics, and
electromagnetics.
Now works on:
Solid earth geophysics; mechanics of earthquakes and
tectonic deformation

**Miloje Makivic**

Computational Research Scientist

Northeast Parallel Architectures Center

111 College Place

Syracuse, New York 13244-4100

miloje@npac.syr.edu

Worked from (1988-end of project):
As graduate student in the Division of Mathematics, Physics and Astronomy at
Caltech, collaborated with the CP group. After 1990, used parallel
resources at CP to develop computational physics algorithms,
specifically Monte Carlo methods on parallel processors for strongly
correlated quantum systems: Spin systems, high-temperature
superconductors, disordered superconducting thin films, and general quantum
critical phenomena. Also worked on self-consistent perturbation theory
approach to heavy fermions and low-dimensional magnets.
Now works on:
From 1990 until September 1993, worked as post-doctoral research in the
Physics Department of Ohio State University. Presently, working at Syracuse
University (NPAC) on the application of parallel computing in industry and
science. Current projects include atmospheric data assimilation and financial
modelling.

**Vincent McKoy**

Professor

California Institute of Technology

Mail Code 127-72

Pasadena, California 91125

bvm@citchem.bitnet

Worked from (3/89-9/89):
Studies of collisions of electrons with polyatomic molecules.
Now works on:
Using variational procedures to obtain cross-sections for electronic
excitation of molecules by electron impact.
Contributed Section 8.3, Studies of
Electron-Molecule Collisions on Distributed-memory Parallel Computers

**Paul Messina**

CCSF Executive Director

California Institute of Technology

Mail Code 158-79

Pasadena, California 91125

messina@CCSF.Caltech.edu

Worked from (1987-end of project):
Involved as Co-Investigator with particular emphasis on acquiring and
managing the computing facilities, and on the systems issues of concurrent
computing environments.
Now works on:
From 1990 until the present, directs the Caltech Concurrent Supercomputing
Facilities, which have pushed to higher limits of performance the approaches
conceived in CP. Also, manages the CASA gigabit network testbed project,
which explores issues on distributed supercomputing.
Contributed Chapter 2, Technical Backdrop

**Steve Otto**

Assistant Professor

Department of Computer Science and Engineering

Oregon Graduate Institute of Science and Technology

20000 NW Walker Rd., P. O. Box 91000

Portland, Oregon 97291-1000

otto@cse.ogi.edu

Worked from (1981-1989):
QCD, Computer chess, fine-grained parallel systems, combinatoric optimization
schemes.
Now works on:
Parallel languages and compilation techniques for scalable parallel
architectures; new combinatoric optimization algorithms.
Contributed Sections 6.6, Character Recognition by
Neural Nets; 7.5, Parallel Random Number Generators; 11.4, An Improved
Method for the Traveling Salesman Problem; 12.7, Sorting; 13.6,
Coherent Parallel C; and 14.3, Computer Chess

**Jean Patterson**

Technical Group Supervisor for

Remote Sensing Analysis Systems and Modeling Group

Jet Propulsion Laboratory

4800 Oak Grove Drive

Mail Code 198-231

Pasadena, California 91109

jep@yosemite.Jpl.Nasa.Gov

Worked from (1984-end of project):
Remote sensing data analysis and modeling for remote sensing applications.
These applications use high-performance parallel processing systems for
the analysis. In particular, involved has been with electromagnetic
scattering and radiation analysis, atmospheric radiative transfer, and
synthetic aperture radar processing.
Now works on:
Continues with electromagnetics and atmospheric radiative transfer
modeling. Key participants involved in the finite element work include
Tom Cwik, Robert Ferraro, Nathan Jacobi, Paulette Liewer, Greg Lyzenga,
and Jay Parker
Contributed Section 9.4, Computational
Electromagnetics

**Francois Pepin**

Staff Member

Canadair Aerospace Group

11324 Meunier

Montreal H3L 2Z6, Canada

Worked from (6/87-end of project):
Simulation of viscous incompressible flows using vortex methods; fast
algorithms for N-body problems.
Now works on:
Simulation of compressible flows over transport aircraft.
Contributed Section 12.5, Fast Vortex Algorithm and
Parallel Computing
**Tom Prince**

Professor of Physics

California Institute of Technology

Mail Code 220-47

Pasadena, California 91125

prince@caltech.edu

Worked from (1985-end of project):
Diffraction-limited imaging with large ground-based optical
and infrared telescopes, very-high sensitivity searches for
pulsars in globular clusters, and searches for pulsars in
short-orbit binary systems.
Now works on:
Very-high speed data acquisition and analysis, image
enhancement of astronomical infrared maps, and pulsar search
and detection.

**Peter Reiher**

Member of the Technical Staff

Jet Propulsion Laboratory

4800 Oak Grove Drive

Mail Stop 525-3660

Pasadena, California 91109

Worked from (11/87-1989):
The TimeWarp operating system, parallel programming synchronization methods.
Now works on:
Parallel and distributed operating systems.
Contributed Section 15.3, Time Warp

**Ken Rose**

Assistant Professor

University of California at Santa Barbara

Department of Electrical and Computer Engineering

Santa Barbara, California 93106

rose@ece.ucsb.edu

Worked from (7/89-end of project):
Combinatin of principles of information theory with tools
from statistical physics for solving hard optimization
problems. Particular applications included fuzzy and hard
clustering (pttern recognition and neural networks), vector
quantization (coding/communications), and tracking.
Now works on:
Information theory (particularly rate-distortion theory),
pattern recognition, source coding, communications, signal
processing, and nonconvex optimization.

**John Salmon**

Research Fellow

California Institute of Technology

Mail Code 206-49

Pasadena, California 91125

johns@ccsf.caltech.edu

Worked from (8/83-end of project):
As a graduate student and post-doc, research interests included
astrophysical applications, Fourier transforms, ray-tracing, parallel
I/O, and operating systems. Still working with the current incarnations of
CP at Caltech.
Now works on:
Fast tree-based methods for N-body problems and other applications
(hydrodynamics, panel methods, random fields) have dominated recent
attention. Work with the current incarnations of CP are still
being continued at present.
Contributed Section 12.4, Tree Codes for N-body
Simulations
**Anthony Skjellum**

Assistant Professor of Computer Science

NSF Engineering Research Center for Computational Field Simulation

and Computer Science Department

Mississippi State University

P. O. Drawer CS

300 Butler Hall

Mississippi State, Mississippi 39762-5623

tony@cs.msstate.edu

Worked from (9/87-end of project):
Parallel libraries, message-passing systems, portability, chemical
engineering applications-flowsheeting.
Now works on:
Same as above, plus standards in message passing (message-passing
interface forum), heterogeneous high-performance clusters
Contributed Sections 9.5, LU Factorization of
Sparse, Unsymmetric Jacobi Matrices; 9.6, Concurrent DASSL Applied to
Dynamic Distillation Column Simulation; and Chapter 16, The Zipcode
Message-passing System

**Michael D. Speight**

Registrar in Medical Radiodiagnosis, Royal Infirmary of Edinburgh

c/o Medical Statistics Unit

University of Edinburgh

Teviot Place

Edinburgh EH8 9AG, Scotland

mds3@edinburgh.ac.uk

Worked from (1989-end of project):
Biologically realistic neural simulation on parallel computers. Most
recent involvement was via Jim Bower's group doing neural simulation
work on the Intel Touchstone Delta.
Now works on:
Virtual reality systems and parallel computing for manipulating medical
images (e.g., human brain MRI scans).
Contributed Section 7.6, Parallel Computing in
Neurobiology: The GENESIS Project
**Eric Van de Velde**

Senior Research Fellow

California Institute of Technology

Mail Code 217-50

Pasadena, California 91125

evdv@ama.caltech.edu

Worked from (6/86-end of project):
Algorithms for concurrent scientific computing; multigrid and linear
algebra algorithms.
Now works on:
Multigrid, linear algebra, fluid flow, reaction-diffusion equations.
Contributed Section 9.7, Adaptive Multigrid
**David Walker**

Member of the Technical Staff

Building 9207A, MS-8083

P.O. Box 2009

Oak Ridge National Laboratory

Oak Ridge, TN 37831-8083

walker@msr.epm.ornl.gov

Worked from (3/86-8/88):
Parallel linear algebra, parallel CFD, benchmarking, programming paradigms,
parallel FFT algorithms.
Now works on:
Linear algebra software for MIMD machines, concurrent particle-in-cell
algorithms for plasma simulations, benchmarking, molecular dynamics.
Contributed Sections 6.2, Convectively-Dominated
Flows and the Flux-Corrected Transport Technique; and 8.1, Full and
Banded Matrix Algorithms

**Brad Werner**

Assistant Professor

University of California at San Diego

Scripps Institution of Oceanography

Center for Coastal Studies

Mail Code 0209

9500 Gilman Drive

La Jolla, California 92093-0209

werner@hayek.ucsd.edu

Worked from (1983-1987):
Simulation of the dynamics of granular materials.
Now works on:
Quantitative geomorphology, nearshore and desert processes, granular
materials, computer simulation, pattern formation.
Contributed Section 9.2, Geomorphology by
Micromechanical Simulations

**Roy Williams**

Senior Staff Scientist

Concurrent Supercomputing Facilities

California Institute of Technology

Mail Code 206-49

Pasadena, California 91125

roy@ccsf.caltech.edu

Worked from (2/86-end of project):
Programming paradigms and algorithms for unstructured triangular meshes.
Now works on:
General unstructured meshes; finite-element and finite-volume methods;
reaction-diffusion equations; mesh generation in complex geometries.
Contributed Chapter 10, DIME Programming
Environment; Sections 11.1, Load Balancing as an Optimization Problem,
12.2, Simulation of the Electrosensory System of the Fish *Gnathonemus petersii*; and 12.3, Transonic Flow

**Carl Winstead**

Assistant Scientist

California Institute of Technology

Mail Code 127-72

Pasadena, California 91125

clw@cco.caltech.edu

Worked from (3/89-9/89):
Computation of electron-molecule collision cross-sections with parallel
machines.
Now works on:
Electron-molecule collision cross-sections relevant to low-temperature
plasmas; improving methods and algorithms in such calculations.

Wed Mar 1 10:19:35 EST 1995