Issues in the architecture of direct interconnection schemes for multiprocessors
Abraham, Seth
This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/20865
Description
Title
Issues in the architecture of direct interconnection schemes for multiprocessors
Author(s)
Abraham, Seth
Issue Date
1990
Department of Study
Computer Science
Discipline
Computer Science
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
Computer Science
Language
eng
Abstract
Direct connected multiprocessors are constructed by connecting processing nodes together with point-to-point links. This thesis will undertake an analysis of issues pertaining to the performance and reliability of direct connected multiprocessor architectures. The hypercube structure is a direct connected network of particular interest because of its simplicity, symmetry, and connectivity. For this reason, the first part of the thesis will focus on the hypercube. Mathematical models, similar to those used for multistage networks, will be developed to predict the performance and reliability of the hypercube. The results of these studies are of practical interest to multiprocessor system architects considering a hypercube interconnection structure. The remainder of the thesis uses the hypercube as a standard for measuring the potential of alternate topologies. First, it examines two direct connection structures that have been considered in the literature to be superior to the hypercube. The results show that these alternate networks fail to deliver the anticipated performance improvements. Second, the thesis introduces a new constraint into the performance analysis of networks--the width of the communication channel between nodes. This study considers the various cost-performance tradeoffs encountered when implementing different network topologies in hardware.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
