| The rapid development of computer technology has brought great changes in geographical information system. With the increasing of spatial data size and computational complexity, geographical information system which uses the centralized model is facing many challenges. P2P technology as a new distributed computing model provides a new way for geographical information sharing. The technology can effectively make use of the edge of network resources and can integrate the network resources of participants. It can improve the efficiency of distributed vector geographical data organization and the performance of spatial indexing by utilizing the storage space and computing capabilities of participants to save amounts of data and process kinds of requests.In this thesis, an optimization mechanism is proposed based on analyzing and summarizing the existing hybrid P2P spatial indexing network. The optimization mechanism is a new idea of group strategy based on the structure of index tree. In this idea, peers with the same spatial data semantics are joined in the same group in which the burden of query is shared by group members together. The current index node in the group is responsible for the management of group members. The group strategy contains node join, the transmission node selection, the current index node replacement and backup and group management mechanism. It can enhance the topology of the original indexing network and improve the query performance and stability of the overall network by using group strategy. The simulation results used PeerSim verify the feasibility of the optimization mechanism. In addition, a P2P based prototype system of dynamic optimization mechanism for spatial vector data indexing network is implemented, which is based on JXTA platform. The experimental results indicate that the indexing network with group strategy can effectively make use of clients'network resources and improve the query performance. |
PDF Full Text Request