Distributed resource search in self-organising networks

Abstract

Virtual Organisations (VO) have emerged as an important field of study in the area of Multi Agent Systems (MAS). A VO consists of a number of members that share preferences or goals with the aim of exploiting the available resources and providing better services than a single member would be able to. In this paper, VOs have been used as a way to cluster heterogeneous agents which have semantically close resources together. A VO-based resource search protocol has been developed with search message routing techniques that have been used to forward resource search messages among VOs instead of individual agents. A decision making component has been added to the agent's body to facilitate the process of maintaining the VO's functionality. Different scenarios have been studied to deal with situations which might affect the VO's work-flow. The proposed solutions have been implemented and tested in a simulated environment. The simulation results have shown a significant improvement in the search results in terms of the quality of matching, the required time to find the requested resources and the success ratio of requests. © 2013 Springer-Verlag.