Optimized Grid Job Scheduling using Hybrid SLFA-GA : A Novel Approach for Hetrogeneous Task

The computing grid is a hardware and software infrastructure that can provide a considerably consistent and cost-effective way for computing. A computational grid is associated with a set of large-scale resources. To manage such sophisticated systems you can use shared resource management tools that try to optimize performance across the entire system. A network planning system is implemented using an efficient algorithm to allocate network resources to the user s program, with the limitations that are required in accordance with the optimization strategy of the requested parameter, and is minimized. In this research, a method for better allocation of resources in the case of hetrogeneous task is proposed . In the proposed method, the genetic algorithm is used to construct and improved the initial population and the shuffled frog-leaping algorithm is used to trace the network and allocate its resources. the main focus of this research is on creating a better initial populationin the hetrogeneous task, which has not been studied in similar works. Meanwhile, it was tried to use genetic algorithm and shuffled frog-leaping algorithms in order to schedule affiliated tasks in work flow model and aim to reduce the end time of tasks. One of the features of this research is that this algorithm finds the right resource for any job in the shortest possible time. The results of implemented in the MATLAB environment showed that in different combinations of heterogeneity, the job and resources of the proposed algorithm show even better results in the number of repetitions.