A multi-agent-based model for service-oriented interaction in a mobile grid computing environment

The paper presents a multi-agent-based model for service-oriented interaction in a mobile grid computing environment. The proposed model consists of three different types of agent: user agents, device resource agents, and device service agents. Device resource agents sell the device resources of the mobile grid through a device resource market. A device service agent makes buying decisions within budget constraints to acquire device resources. A service-oriented multi-agent interaction algorithm is proposed. In this algorithm, an initial set of prices is announced to the user agent, and the users make their decisions based on the device service agent’s price policy, and iteratively approach an optimal solution. Device service agents wish to buy resources cheaply from device resource agents and sell device services at a profit to user agents. Device service agents try to maximize their profits by adjusting the supplier role behavior on the basis of the feedback they receive from the service market; at the same time, they try to buy the device resource with lowest cost through the resource market. In the simulation, a performance evaluation of the algorithm is conducted.