Energy-constrained Dynamic Scheduling And Dynamic Pricing Algorithm In Cloud Computing

Since the concept of cloud computing was proposed, cloud computing rapidly became a core technology of the communications network, and gradually formed a new business service model. With the deepening and development of cloud computing technology, cloud computing gradually change from theory to practice. Many IT companies have started research on cloud computing technology, and produced data center which has the commercial wireless cloud service capabilities. According to studies, the energy consumption of data centers has become an important part of the energy consumption. Thus, in the increasingly competitive market of cloud services, it has become an urgent problem that how to maximize profit of the cloud service providers under the limited energy consumption.Based on the above problems, there have been many algorithms to maximize the profits of the cloud service providers in recent years. However, these algorithms always focus on a partial aspects of how to maximize the cloud service providers’ profits during the actual implementation, or some algorithms only consider dynamic scheduling or dynamic pricing for the cloud resources, or some algorithms only consider the limited energy conditions of cloud data center, etc. However, the above mentioned issues must be resolved when optimize the profits of the cloud service providers.Therefore, in this paper an energy-constrained dynamic scheduling and dynamic pricing algorithm is proposed. The algorithm not only satisfy the limited energy consumption but also consider dynamic scheduling and dynamic pricing for the cloud resources, so that the long-term profit of the cloud service operators could be maximized cloud service providers. At the same time, this paper pulled Lyapunov optimization theory in the process of optimization algorithm. This made dynamic scheduling and dynamic pricing to be realized and further optimized the cloud service provider’s profits. Depending on the system model, the profit optimization algorithm that was proposed in this thesis can be divided into two different types:(a) Energy-constrained dynamic scheduling and dynamic pricing algorithm.(b) Different QoS requirements of energy-constrained dynamic scheduling and dynamic pricing algorithm. The two algorithms used the Lyapunov optimization theory during the realization. Finally, through simulation and data analysis, the algorithm can be compromised between profit and delay sequence while meeting the requirements of energy restriction.