Improving the Grid Scheduling Performance with Fault Tolerance Using Genetic Algorithm

In the last few decades we have witnessed the emergence of grid computing as an innovative extension to distributed computing technology, for computing resource sharing among participants in a virtualized collection of organizations. Grid computing entails new challenges as the adaptation of heterogeneous resources unlike homogeneous resources cluster in distributed systems. It is important to maintain proportional fairness in the grid scheduling in order to achieve balanced scheduling. In this paper we propose the importance of genetic algorithm to design schedulers that minimizes the waiting time and maximizes the resource utilization and provides fairness in the grid environment. The resource types and their efficiency are considered in order to maximize their utilization. This paper proposes a solution to maximize the throughput while considering multiple job requests during the scheduling process. The idea of fault tolerance in the crash fault environment will also be implemented based on precautionary method and real time restoration.