Jianguo Cai

Date of Award

4-1-1999

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Dr. Hesham H. Ali

Second Advisor

Dr. Heshem El-Rewini

Abstract

A scheduling problem is an issue that needs to be addressed whenever a need is present to arrange a set of tasks into a set of processing units, under certain policies, with different possible outcomes. In general, the time complexity of finding an optimal solution for scheduling problem is exponential. However, in many situations, finding an optimal solution for a scheduling problem is essential. This necessity for efficient scheduling has spurred much research work in this area. As a result, many efficient heuristic scheduling algorithms have been developed. For the most part, however, these scheduling algorithms apply to parallel programs, the algorithms themselves are sequential and, as yet, little work has been done on paralleling scheduling algorithms. In this thesis, we study the issue of parallel scheduling and present a new parallel scheduling scheme. The primary objective is to study parallel scheduling algorithms by comparing their performances with sequential scheduling algorithms. In this study, horizontal scheme, vertical scheme, as well as a new scheme, hybrid scheme, are implemented, and compared via simulation. The results of the conducted experiments show that horizontal scheme algorithms normally produce shorter schedules, while the vertical scheme algorithms have better speedups. It also shows that the hybrid scheme achieves better parallelism, while still producing acceptable schedule length by producing schedules with the advantages of both horizontal and vertical schemes.

Comments

A Thesis 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 1999 Jianguo Cai

Download

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS