Task Scheduling Algorithm Based on the Genetic Algorithm and Dynamic Adaptive Scheduling in a Cloud Computing Environment

Cloud computing is a concept introducing in the world of information technology recently, and provides an environment for sharing sources. In this environment, an efficient and effective scheduling system plays a fundamental role. Hence, the problem of scheduling tasks in cloud computing is a very important issue, which tries to identify an optimal scheduling for performing tasks and allocating an optimal source. The purpose of this research is to provide a new hybrid model for optimization of scheduling tasks based on combination of dynamic adaptive and genetic algorithms in a cloud computing environment. The hybrid model aimed to cover the defects of these algorithms. Based on this model, an initial solution was presented for resolving the weakness of the genetic algorithm, the action of low speed as well as the weakness of the dynamic adaptive algorithm, trapping in local answers. In the proposed model, first, dynamic adaptive algorithm could find an appropriate solution for scheduling problem within a good time using the technique of searching required resources by KD tree and data retrieval in parallel. This solution, as an initial answer, was given to the genetic algorithm, and this algorithm performed the optimal research. As a consequence, the final algorithm, a combination of these two algorithms, provides a better load balancing for cloud resources in a relatively acceptable time.