Research Of Resource Discovery Method In Grid Environment

With the growing sharing requirements of people to the wide area network resource day by day, the grid technology has gradually become one of the major forms of achieving the resource sharing in the internet environment. In order to solve the problem of grid resource discovery, a grid resource discovery method of adjustment which utilizes preference partition and M-Flooding algorithm is proposed in this paper after analyzing the shortcomings of the traditional method of measuring resource similarity degree and the Flooding message dissemination. First, this method gives a mew method of measuring resource similarity degree, the grid spatial nodes are divided into different groups according to their respective preference attributes. Searching requests are transmitted in groups, thus avoiding the disadvantages brought by the traditional blind search. Second, this method gives an improved message diffusion way M-Flooding algorithm, reducing the message dissemination cost and redundant messages transmission cost.In order to go further to solve the shortcomings of traditional resource discovery mechanism, a grid resource discovery model based on improved architecture and multicast is proposed in this paper. Sophisticated routing strategies are implemented to provide better scalability and routing efficiency in this paper. Nodes are divided into different domains, improved distributed hash table technologies are adopted within domains to achieve better local routing efficiency. The combination of network coverage and hierarchical tree structure improves the overall scalability; The messages are transmitted among the management nodes through multicast, reducing the network load.The grid resource discovery method of adjustment which utilizes preference partition and M-Flooding algorithm and the grid resource discovery model based on improved architecture and multicast are simulated using the simulation tool GridSim, comparing the two method with traditional grid resource discovery method:The grid resource discovery method of adjustment which utilizes preference partition and M-Flooding algorithm improves the efficiency of grid resource discovery and reduces the average path length of resource discovery comparing with the traditional grid resource discovery method Gnutella. The grid resource discovery model based on improved architecture and multicast improves the resource search success rate in grid environment comparing with the Flooding resource discovery model based on the fully distributed way. We find that we can reduce the number of query and response transmitted over the network by means of limiting hops, as well as the query time is minimized and the efficiency of resource discovery is improved in grid environment through comparing the change of performance parameter in different TTL.