Mobile agents and energy-efficient multidatabase design

Multidatabase systems are widely used to integrate distributed heterogeneous data sources. To support mobile data access, existing multidatabases must address the obstacles in mobile computing such as low bandwidth, intermittent network connections, and resource constraints of mobile devices. The execution autonomy of mobile agents alleviates these problems and provides opportunity for conserving energy. We proposed and implemented a novel multidatabase information retrieval system, MAMDAS, which represents an application of Mobile Agents within the Mobile Data Access System framework. In this paper, we propose several MAMDAS optimization techniques to improve its performance and study the effect of mobile agents on energy saving through extensive simulation. Our experimental results show that by deploying appropriate power management policies, the mobile device energy saving can achieve as high as 51%. The high performance, energy efficiency, scalability, and robustness of MAMDAS allow us to envision its application in a variety of fields such as military operations, emergency teams, and mobile e-business.