IP-paging based Broker Service for Reliability and Efficiency in Mobile Grid Computing

Abstract

A mobile grid is emerging as a new opportunity to integrate and utilize the exploding population of mobile resources. Mobile resources such as laptops, PDAs, and embedded devices are integrated in the mobile grid. In mobile grid, a mobile device can act both as a service requester utilizing grid resources and as a service provider lending its capability to the grid anytime anywhere. In this paper, we focused on the usage of mobile devices as resources. Therefore, we proposed the mobile grid computing architecture based on IP-paging. To provide the reliability in mobile grid computing,, the proposed architecture must overcome several weak points of wireless networks. First of all, to find idle mobile device and collect mobile resource status information, the resource management scheme based on IP-paging scheme is proposed. Although the idle mobile resources are secured, mobile resources still have the disconnection problem and the power capacity problem by property of mobile device. To solve these problems, we propose task migration scheme using user-defined checkpoint. Proposed task migration scheme determines the point of checkpointing and decides task migration by analysis of wireless signaling strength and power capacity. And, to minimize application complete time, we propose scheduling algorithm based on power capacity. It strives to minimize the application complete time and prevent re-execution. By performance evaluation, proposed mobile gird architecture provide the reliability of mobile resource, the robustness of execution, and the minimization of application complete time.