Abstract

With the rapid development of high-speed wide-area networks and powerful yet low-cost computational resources, grid computing has emerged as an attractive computing paradigm. The computational grid is a new parallel and distributed computing paradigm that provides resources for large scientific computing applications. The main techniques that are most suitable to cope with the dynamic nature of the grid are the effective utilization of grid resources and the distribution of application load among multiple resources in a grid environment. This paper addresses the problem of scheduling and load balancing in a grid environment. A Decentralized Dynamic load balancing algorithm is proposed which combines the strong points of neighbor based and cluster based load balancing techniques. This algorithm estimates system parameters such as resource processing capacity, load on each resource and transfer delay for scheduling and load balancing. A set of simulation experiments show that the proposed algorithm provides significant performance over existing ones.