中文摘要：

在云计算环境下,租户向云服务提供商租赁云产品和服务,那么云服务商如何规定一个合理的服务定价使得多租户与提供商都满意成为一个亟待解决的问题.针对该问题,该文提出一种基于帕累托最优思想的服务定价模型,综合考虑多租户与提供商的利益,使用多目标粒子群优化算法得到全局最优的资源分配与服务定价结果.由于迭代算法的效率与服务定价实时性的需求存在冲突,所以该文提出两阶段定价策略：独立定价与集中定价.独立定价阶段,参考历史同需求或相似需求的定价,结合当前资源使用占比给出实时报价;集中定价以一定时间为周期,根据已知的该周期内多租户提出的不同需求,提前使用粒子群算法求出最优服务定价与资源分配策略.实验表明,通过该文提出的定价模型,可以得到一个使多租户和云服务提供商都满意的定价,并且在定价过程中,采用隐私保护技术,有效地保护了租户的数据安全.

英文摘要：

In the circumstance of cloud computing,tenants lease cloud products and services from cloud service provider.How to give a reasonable price satisfying both provider and tenants is a challenge for provider.This paper proposes a pricing model based on Pareto Optimization thought with application of Multi-Objective Particle Swarm Optimization Algorithm.It gives a global optimal resource allocation and price result,considering comprehensively profits of both provider and tenants.In order to avoid conflict between iterative algorithms＇ low efficiency and the service pricing＇s real-time demand,the thesis separates the pricing process into two stages,referring to Solitary Pricing and Centralized Pricing.In the Solitary Pricing stage,we give a real-time price referring to the history data with the same demand or similar demand and current resource usage ratio.In the Centralized Pricing stage,we give an optimal price and resource allocation strategy with Pareto Optimization thought based on different known demands proposed by multi-tenants in a certain period.The experimental results show that the proposed pricing model can get a price satisfying both provider and tenants,and in the pricing process,using the privacy protection technology,effectively protects the data security of tenants.