An Improved Genetic-Based Approach to Task Scheduling in Inter-cloud Environment

With the development of cloud computing, the number of cloud computing service providers has arisen rapidly. The research of task scheduling in cloud computing environment nearly enters a mature stage. But when there is a sharp increase in the amount of user tasks, a single cloud provider cannot meet user's needs. This phenomenon prompted the generation of Inter-cloud, and the task scheduling in which gradually gets everyone's attention. In this paper, we improve the genetic algorithm by adopting Gene Space Balance Strategy (GSBS), which optimizes the generation of initial population. On the basis of improved algorithm, we propose the multi-objective optimization task scheduling method in Inter-cloud. The scheduling goal is to minimize the completion time and cost. We can complete task scheduling according to the different QoS requirements of users. By performing simulation on cloud Sim, we demonstrate the effectiveness of improved algorithm. At the same time, we compare the scheduling results of single-cloud and Inter-cloud.