Title: Load balancing consideration of both transmission and process responding time for multi-task assignment

Authors: Yean-Fu Wen; Chih-Lung Chang

Addresses: Graduate Institute of Information Management, National Taipei University, 151, University Rd., San Shia District, New Taipei City 23741, Taiwan ' Graduate Institute of Information Management, National Taipei University, 151, University Rd., San Shia District, New Taipei City 23741, Taiwan

Abstract: The existing load balancing research mainly the unilateral fairness of the transmission network or only stand for the processing host to handle fairness issue. This work addresses the load balancing between network and host to assign tasks to the related hosts along the path with minimal processing and transmission delays subject to the capacity of processing hosts and transmission links. Three task schedulers: first-come-first-service (FCFS), minimum task assign to minimum completion time (Min-Min), and the proposed minimised the maximal process and transmission time (Min-Max) allocation mechanisms, are adopted along with biased random sampling (BRS) mechanism to observe the distribution of interactive results. This study changes three variables manipulated with the number of hosts, the number of tasks, and the expected processing time and records the maximal/average responding time and fairness index, to analyse the load balancing results. The results show that the Min-Max and BRS of the match that is used to obtain the lowest responding time to satisfy the requirement of users.

Keywords: cloud computing; heterogeneous networks; load balancing; responding time; situated computing; task assignment; multi-task assignment; task scheduling.

DOI: 10.1504/IJSSC.2014.062470

International Journal of Space-Based and Situated Computing, 2014 Vol.4 No.2, pp.100 - 113

Received: 06 Jan 2014
Accepted: 09 Jan 2014
Published online: 05 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article