Abstract
Major problem which affects the Grid processing is its network is its bandwidth which differs from Grid network to network. In order to rectify this problem, very few algorithms have been suggested earlier keeping in view of computing grid resources based on the concept of load balancing. None of the existing algorithms has been found experimentally suitable in order to overcome the drawbacks present in the grid computing system. In this article, we propose a novel architecture to overcome all the disadvantages present in the Grid computing architecture. Results obtained from the experiments suggested that the proposed architecture was of great use in balancing the load and allocating to the resources to the groups of jobs using Grouping Based Dynamic Job Scheduling algorithm. Our proposed algorithm is compared with First Come First Served Job Scheduling algorithm (FCFS) and Ant Colony Optimization Job Scheduling algorithm (ACO).