Browse Prior Art Database

CALTECH CONCURRENT COMPUTATION PROGRAM A Campus/JPL Collaboration TECHNICAL BULLETIN

IP.com Disclosure Number: IPCOM000127956D
Original Publication Date: 1987-Dec-31
Included in the Prior Art Database: 2005-Sep-14
Document File: 17 page(s) / 64K

Publishing Venue

Software Patent Institute

Related People

Neal Larsen Palmer: AUTHOR [+4]

Abstract

CONTENTS

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 5% of the total text.

Page 1 of 17

THIS DOCUMENT IS AN APPROXIMATE REPRESENTATION OF THE ORIGINAL.

CALTECH CONCURRENT COMPUTATION PROGRAM A Campus/JPL Collaboration TECHNICAL BULLETIN

No. 11, September, 1987

CONTENTS

1. News Items * Enhancements to NCUBE * XENIX on PC NCUBEs * New Personnel * Visitor from Toronto * Major Donation of Chips by Motorola * Collaboration with Rice University

2. Projects Under Development . * Free-Lagrange Hydrodynamics in Parallel

3. Recent Talks . . . . . . . . . _ _ _ _ _ _ _ _ _

:Multitarget Tracking with a Hypercube" '"'Synthetic Aperture Radar on a Hypercube" * "Distributed Software Architecture: The Time Warp Example" * "The Transputer and OCCAM" * " 3D Vortex Dynamics on a Cray X/MP and Intel iPSC" * "Comparing Shared and Distributed Memory Computers"

4. Recent C3P Reports and Publications . . . . . . . . . .

* Benchmarking the Connection Machine * Emerging Supercomputer Architectures * The Parallel Solution of Nonlinear Elliptic Equations * Time-Dependent Hartree-Fock Calculations on the Hypercube * A Unification of Approaches to Shared, Hierarchial and Distributed Memory Machines 5. A Comparison of Shared and Distributed Memory Computers * CrOS on the BBN Butterfly * The Balance Multiprocessor: Part 0. Hardware

8. Articles of Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20

* Three Supercomputer Conferences * Two Supercomputer Performance Hypercubes

The Caltech Concurrent Computation Program Technical Bulletin reports current developments and items of interest to developers and users of hypercube concurrent processor technology at Caltech and JPL. To subscribe to or receive issues of the C3P Technical Bulletin, write to: Caltech Concurrent Computation Program Technical Bulletin c/o Geoffrey Fox Mailcode 158-79 Caltech Pasadena, CA91125

Editors: Neal Larsen Palmer, David W. Walker

Contributors: Clive Baillie, Geoffrey C. Fox, Roy Williams

***~* 2

1. News Items

California Institute of Technology Page 1 Dec 31, 1987

Page 2 of 17

CALTECH CONCURRENT COMPUTATION PROGRAM A Campus/JPL Collaboration TECHNICAL BULLETIN

* Enhancements to NCUBE The C3P NCUBE now has 512 nodes, each with 0.5 Mbytes of memory. It is not currently intended to expand the NCUBE to its maximum capacity of 1024 nodes. Work on enhancing the NCUBE is now concentrating on three main areas: (1) the use of the parallel graphics interface, (2) optimizing internode communication by means of a set of fast CrOS routines, and (3) improving the software environment. A report on progress in these areas will be made in a subsequent issue of the Technical Bulletin.

* XENIX on PC NCUBEs The XENIX operating System has now been ported to PC NCUBEs, for use in place of AXIS. Users interested in implementing XENIX on their PC NCUBE should contact Adam Kolawa (ext. 2907.

* New Personnel Two new scientists, Jeff Koller, and Hong-Quang Ding, have recently joined the C3P group. Jeff Koller joined C3P at the beginning of July as a post dot, after a 3 year post dot a...