Date of Award

Fall 1994

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Electrical & Computer Engineering

Program/Concentration

Computer Engineering

Committee Director

James F. Leathrum, Jr.

Committee Member

John W. Stoughton

Committee Member

Martin Meyer

Call Number for Print

Special Collections LD4331.E55 N64

Abstract

Locally shared memory systems offer significant advantages over other parallel processing systems for specific classes of problems. Locally shared memory systems tend to be easier to program because explicit passing of messages is not necessary. The goal in defining a locally shared memory system is to allow only a small number of processors access to any single memory. If this goal is met, locally shared memory systems provide an architecture which is relatively simple to implement.

To make shared memory architectures attractive to designers of special purpose parallel architectures, the architectures must be scalable. To be scalable, the number of processors connected to each locally shared memory must be relatively small and remain fixed or grow slowly as the number of processors grows. For classes of algorithms where the data dependencies tend to be near neighbor, locally shared memory architectures may be defined which are scalable.

The methodology developed in this thesis helps the designer of special purpose architectures by defining constructs for locally shared memories. To define these constructs, data dependencies are examined, and data are assigned to memories based on those dependencies. The interconnection of processors and locally shared memories are then defined based on the data dependencies. The methodology reduces the connectivity of the locally shared memories by intelligently mapping data into memory modules. The methodology is demonstrated for a multigrid algorithm, an N-body problem solved using the direct method, and an N-body problem solved using the Fast Multipole algorithm.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

10.25777/cafk-fq93

Download

Share

COinS