Email RIS file
Bookmark
Report an Issue

A structured representation for parallel algorithm design on multicomputers [electronic resource].

Published
Washington, D.C : United States. Dept. of Energy. Office of Energy Research, 1991.
Oak Ridge, Tenn. : Distributed by the Office of Scientific and Technical Information, U.S. Dept. of Energy.
Physical Description
Pages: (7 pages) : digital, PDF file
Additional Creators
Ames Laboratory, United States. Department of Energy. Office of Energy Research, and United States. Department of Energy. Office of Scientific and Technical Information
Access Online

Availability

I Want It
I Want It

Finding items...

Restrictions on Access
Free-to-read Unrestricted online access
Summary
Traditionally, parallel algorithms have been designed by brute force methods and fine-tuned on each architecture to achieve high performance. Rather than studying the design case by case, a systematic approach is proposed. A notation is first developed. Using this notation, most of the frequently used scientific and engineering applications can be presented by simple formulas. The formulas constitute the structured representation of the corresponding applications. The structured representation is simple, adequate and easy to understand. They also contain sufficient information about uneven allocation and communication latency degradations. With the structured representation, applications can be compared, classified and partitioned. Some of the basic building blocks, called computation models, of frequently used applications are identified and studied. Most applications are combinations of some computation models. The structured representation relates general applications to computation models. Studying computation models leads to a guideline for efficient parallel algorithm design for general applications. 6 refs., 7 figs.
Report Numbers
E 1.99:is-m-657
E 1.99: conf-9104158--1
conf-9104158--1
is-m-657
Subject(s)
Other Subject(s)
Note
Published through SciTech Connect.
01/01/1991.
"is-m-657"
" conf-9104158--1"
"DE91006674"
6. distributed memory computing conference, Portland, OR (USA), 28 Apr - 2 May 1991.
Ni, L.M. . Dept. of Computer Science; Sun, Xian-He.
Funding Information
W-7405-ENG-82
View MARC record | catkey: 14354448