Author

Haiying Sun

Date of Award

10-1-2003

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Dr. Hesham H. Ali

Abstract

The problem of scheduling tasks across distributed system has been approved to be NP-complete in its general case. When communication cost among system processors is not considered, polynominal-time optimal algorithms for solving scheduling problem are exit only in three special cases. In attempting to solve the problem in the general case, a number of heuristics have been developed. These algorithms intend to reduce the input task graph to one of the special cases and then optimal scheduling can be obtained accordingly. In this paper, we study all these heuristics, and present a improved heuristic --- “Maximal Tree graph approach for scheduling general task graph in the parallel system. A package is developed for comparing the proposed heuristic with three other algorithms, List, Maximal chain and Augmentation. A number of experimental studies have been conducted to compare the proposed technique with these known heuristics. Finally, the conclusion of the algorithm is most efficient for a certain kind of task graph was made accordingly.

Comments

A Thesis-Equivalent Project Presented to the Department of Computer Science and the Faculty of the Graduate College University of Nebraska In Partial Fulfillment of the Requirements for the Degree Master of Science University of Nebraska at Omaha. Copyright 2003 Haiying Sun

Download

COinS