A pareto-based Artificial Bee Colony and product line for optimizing scheduling of VM on cloud computing

In this paper, we present a task scheduling management based on the utility model which is used in economics to represent the needs of both the client and the provider. Indeed, our work copes with two man parameters that affect the broker, the cost of virtual machine instances and their response time. Minimizing those two objectives give the best quality of service to the customers and offer the broker an important profit. In fact, we consider the virtual machines as a product line and use the feature models to represent the virtual machines configurations to select the efficient resources that suit customer requirements and try at same time to minimize virtual machine cost. An efficient task scheduling mechanism can not only fit client's requirements, but also improve the resource utilization, be aware of the changing environment and intends to try to balance the system. Thus, our work is based on Artificial Bee Colony to optimize the scheduling of tasks on virtual machine in cloud computing by analyzing the difference of virtual machine load balancing algorithm.