An Application of BP Neural Networks Based on Adaptive Genetic Algorithm for Load Balancing in VOD Cluster

摘要

In view of BP neural network is simple and good characteristics of plasticity, and genetic algorithm is the inherent implicit parallel, better global optimization capability, adaptability and so on, established a BP neural network based on adaptive genetic algorithm for load balancing in VOD cluster. The testing results,compared with weighted round robin algorithm and weighted least connections algorithm, indicate that this algorithm reduces processor average time to complete the task, and increases the number of throughput and connectivity, and improve the handling capacity of the VOD server at no additional server or improving server configuration.