A New Load Balancing Method For Distributed Memory Multicomputers

The load balancing algorithms are based on the intuition that for better resource utilization it is desirable for the load in distributed system to be balanced evenly. Thus, a load balancing algorithm tries to distribute the total system load among the processing elements such that each processing elements receives approximately equal amount of load. A novel distributed algorithm for load balancing is proposed here which is based on the global load balancing information and current load distribution of the system. A simulation model is developed to compare the performance of the proposed method with the previously stated methods like MCSTLB (Maximum Cost Spanning Tree Load Balancing) method, BTLB (Binary Tree Load Balancing) method, and CBTLB (Condensed Binary Tree Load Balancing) method and Cluster method. Two criteria, the execution time and the number of process migration required by different load balancing methods are used for performance evaluation. The experimental result shows that the execution time and the number of process migration required by the proposed method is better than that of the existing methods.