Skip to Main Content
This paper analyses the defect of the designing model of load balancing in hybrid P2P network, and proposes a new strategy of scheduling for load balancing, which based on culture algorithm of multilayer belief spaces. According to the current load status of each super-node in the system, it can calculate its load weight of the next time section by using the cultural algorithm, and assign the current requests of login with weighted round-robin. The extracted knowledge which is got by selecting the best belief space from the multilayer belief space is used to optimize the constrains of the evolutionary, in order to accelerate the optimization process of the calculation of load weight. Simulation experiments show that the allocation of login request to each node is to be balance much faster through using of cultural algorithm based on multilayer belief spaces.