On Balancing Workload in a Highly Mobile Environment

Jeffrey Xu Yu
Department of Computer Science Australian National University Canberra, ACT 0200, Australia yu@cs.anu.edu.au

Kian-Lee Tan
Dept. of Information Systems & Computer Science National University of Singapore Lower Kent Ridge, Singapore 119260 tankl@iscs.nus.sg

Xun Qu
Computer Sciences Laboratory Australian National University Canberra, ACT 0200, Australia quxun@cslab.anu.edu.au


Traditionally, in a distributed database system, it has been assumed that the data requirements of most queries can be met at the site where the query is submitted. However, this concept of localization is challenged by new technologies that facilitate mobility of users. In a highly mobile environment, queries issued by a mobile user are more likely to require remote access. As a result, existing algorithms may no longer provide desirable performance. In this paper, we propose and study several algorithms for improving the throughput of the system in such a context. Our simulation results show that unless the communication bandwidth is low, strategies that balance the workload based on the machine load perform best.
Conference Home Page