Authors: Youcef Derbal
Addresses: Ted Rogers School of Information Technology Management, Ryerson University, 350 Victoria Street, Toronto, Ontario, M5B 2K3, Canada
Abstract: Grids are geographically distributed aggregates of resource nodes that support the provision of computing services, including computing cycle, simulation services, data mining and data processing services. Grids span multiple management domains with different service provisioning strategies. In this paper, we present a decentralised strategy of service discovery that utilises a selective service capacity state dissemination and an experience-based confidence model. The model provides a measure of the likelihood that a discovery request forwarded to a peer node would lead to a match between the requested capacity and the node|s available service capacity. The simulation results show that the proposed algorithm outperforms both the Flooding as well as the random forwarding discovery schemes.
Keywords: service discovery; service capacity state dissemination; confidence models; neighbourhood topology; computational grids; simulation; grid services.
International Journal of Web and Grid Services, 2008 Vol.4 No.2, pp.189 - 210
Published online: 21 Jun 2008 *Full-text access for editors Access for subscribers Purchase this article Comment on this article