云市场下云服务商与终端用户间利益均衡的优化模型与算法研究

With the growing number of successful applications of Cloud Computing Services, many researchers have devoted themselves into the economic benefit problems between the end-users, SaaS providers, PaaS providers and IaaS providers in the cloud service system. However, little attentions are paid on researching the benefit game among the participants in the cloud market to make all of them to obtain the win-win strategy from the perspective of the overall revenue of the cloud services system. The main contents of the project are as follows: (1) Based on the structural characteristics of the hierarchical optimization, we analyze the nested hierarchical game relationship among the participants in the cloud service system. Then we establish the following three models: the cloud service pricing model with end-users' participating the decision-making process, the efficient allocation model of cloud services under the contract mechanism and spot market and the model framework of the optimal strategy of the cloud services composition. Through these models, we want to research how to make each cloud providers and end-users to obtain the best benefit.(2) The intuitionistic fuzzy interactive algorithm will be proposed for solving the hierarchical optimization problems to obtain the optimal win-win strategy. In addition, we will propose the effective numerical optimization algorithms and intelligent algorithms to solve the other hierarchical optimization problems. (3) The certain numerical experiments will be implemented to verify the correctness of the model and the effectiveness of the algorithm. This research has important theoretical significance and application value.

随着越来越多云服务的成功应用，云市场下终端用户、SaaS服务商、PaaS服务商和IaaS服务商所构成的云服务系统中各方间的经济利益问题受到众多学者的广泛关注与深入研究。然而从云服务系统整体收益角度出发讨论云市场下各参与者间的利益博弈问题使得所有参与者能够获得共赢的最优策略却鲜有研究。本项目的主要研究内容为：(1)基于层次优化的结构特点，分析云服务系统中各参与者之间的嵌套递阶博弈关系，建立终端用户参与决策的云服务定价策略模型，云代理(Broker)合约机制和现货市场下云服务资源有效分配模型及最优云服务组合框架，使得各个云服务商和终端用户获得最佳收益。(2) 重点研究求解层次优化模型的直觉模糊交互式算法，得到使得各层决策者获得共赢的最优策略，并构造层次优化问题有效的数值优化算法和智能算法。(3)进行数值模拟计算，验证模型的正确性和算法的有效性。本项目的研究具有重要的理论意义和应用价值。

云市场中由终端用户、SaaS 服务商、PaaS 服务商和IaaS 服务商所构成的云服务系统具有明显的递阶嵌套层次关系，建立相关层次优化模型使各方决策者获得利益均衡具有重要的应用价值。本项目主要研究内容及成果为：.(1) 云计算定价问题的研究。为了尽可能减少决策者主观因素的影响并考虑到环境中的不确定性，结合直觉模糊理论，用隶属函数表示决策者对决策结果的接受程度，用非隶属函数表示决策者对决策结果的拒绝程度。由隶属函数和非隶属函数构造得分函数来刻画决策者对所得解的满意程度，最后通过构造直觉模糊交互式方法得到了既体现了决策者的主观意识又符合客观事实的满意解，给出了使得各层决策者获得共赢的决策策略。.(2) 层次优化问题的相关理论与算法。针对双层多目标规划问题、单领导多随从问题、多领导随从问题、弱线性双层规划问题和双层变分不等式，得到了其最优性条件并构造有效的数值优化算法和智能算法。尤其是针对下层不满足Slater约束规格的双层规划问题，通过对下层问题进行扰动，得到了原双层规划问题的近似解。.(3) 均衡约束问题的研究。引进新的约束规格并给出了其M-稳定点的最优性必要条件。探讨了具有约束的向量变分不等式的可分与择一问题，利用图像空间分析方法得到了具有约束条件的向量变分不等式及其连续性的双间隙函数。另一方面，构造了几种有效的求解均衡问题的算法，如邻近点算法、非点同论光滑算法和合成投影法。.(4)其他优化问题的研究。如锥优化的最优性条件与求解算法、稀疏优化的算法设计和多目标优化与变分不等式理论研究等等。