A new scheduling algorithm for dynamic task and fault tolerant in heterogeneous grid systems using Genetic Algorithm

In this paper with studying of all parameters in grid environment a new scheduling algorithm for independent task is introduced according to Genetic Algorithm. This algorithm can be more efficient and more dependable than similar previous algorithms. The simulated results and reasons for reaching to better makespan and more efficiency in the grid environment. In the grids with high fault with high fault rate for fault tolerant is used from check point method that has more efficiency that other methods such as retry, migration and replication. This method maintains effective efficiency in these situations. So the servicing quality increases in various grid environments and also the average time of task recoveries decreases considerably The main purpose of this paper is reducing the repeating of the generations in Genetic Algorithm for reaching higher speed and also considering the communications costs (available in fitness function) with maintaining the fitness efficiency. The simulations are done with Gridsim for showing the created improvement at proposed algorithm rather than previous algorithms.