Welcome to the InfoSci Platform
Reference Hub3
A Dynamic Load Balancing Strategy with Adaptive Thresholds (DLBAT) for Parallel Computing System

A Dynamic Load Balancing Strategy with Adaptive Thresholds (DLBAT) for Parallel Computing System

Taj Alam, Zahid Raza
Copyright: © 2014 |Volume: 5 |Issue: 1 |Pages: 16
ISSN: 1947-3532|EISSN: 1947-3540|EISBN13: 9781466653672|DOI: 10.4018/ijdst.2014010104
Cite Article Cite Article

MLA

Alam, Taj, and Zahid Raza. "A Dynamic Load Balancing Strategy with Adaptive Thresholds (DLBAT) for Parallel Computing System." IJDST vol.5, no.1 2014: pp.54-69. http://doi.org/10.4018/ijdst.2014010104

APA

Alam, T. & Raza, Z. (2014). A Dynamic Load Balancing Strategy with Adaptive Thresholds (DLBAT) for Parallel Computing System. International Journal of Distributed Systems and Technologies (IJDST), 5(1), 54-69. http://doi.org/10.4018/ijdst.2014010104

Chicago

Alam, Taj, and Zahid Raza. "A Dynamic Load Balancing Strategy with Adaptive Thresholds (DLBAT) for Parallel Computing System," International Journal of Distributed Systems and Technologies (IJDST) 5, no.1: 54-69. http://doi.org/10.4018/ijdst.2014010104

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The primary objective of scheduling is to minimize the job execution time and maximize the resource utilization. Scheduling of ‘m' jobs to ‘n' resources with the objective to optimize the QoS parameters has been proven to be NP-hard problem. Two broad approaches that are defined for dealing with NP-hard problems are approximate and heuristic approach. In this paper, a centralized dynamic load balancing strategy using adaptive thresholds has been proposed for a multiprocessors system. The scheduler continuously monitors the load on the system and takes corrective measures as the load changes. The threshold values considered are adaptive in nature and are readjusted to suite the changing load on the system according to the mean of the available load. Effectively, the load is leveraged towards the mean, transferring only the appropriate number of jobs from heavily loaded nodes to lightly loaded nodes. In addition, the threshold values are designed in such a way that the scheduler avoids excessive load balancing. Therefore, the scheduler always ensures a uniform distribution of the load on the processing elements with dynamic load environment. Simulation study reveals the effectiveness of the model under various conditions.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.