摘要
云市场用户的资源需求往往会随着时间而波动变化,在资源分配与定价时若不充分考虑供需双方的内在激励,将难以获得理想的结果。基于市场策略,设计一个组合拍卖机制来平滑用户需求,以提高资源管理效率及服务收益;提出的机制以动态定价的方式向用户分配资源,实现了无妒与可信两种属性,无妒属性保证了机制运行的稳定性,而可信属性可以使得服务收益最大化。此外,提出的方法具有较低的计算复杂度,易于实现。实验结果表明,在短缺与饱和市场下,所提方法均可获得近似最优的收益及相对较高的社会福利。
In the cloud market,the computing resource requirement of users usually fluctuates with the time change,and it might achieve undesired outcomes while carelessly considering the incentive of both the server and the user in such scenario.This paper investigated how to design a combinatorial auction mechanism to smooth user demand for improving resource ma-nagement and maximizing service revenue.The proposed mechanism allocated resources to users in a dynamic pricing method,which realized two attributes called envy-freeness and truthfulness.The envy-freeness ensured the stability of mechanism operation,while the truthfulness could maximize the service revenue.In addition,the proposed method has low computational complexity and is easy to be implemented.The experimental results show that the proposed method can obtain approximately optimal revenue and relatively high social welfare in both short and saturated markets.
作者
杨波
刘文彬
杨涵
钟敏娟
屈喜龙
Yang Bo;Liu Wenbin;Yang Han;Zhong Minjuan;Qu Xilong(College of Information&Management,Hunan University of Finance&Economics,Changsha 410205,China;School of Information&Electronics,Beijing Institute of Technology,Beijing 100081,China)
出处
《计算机应用研究》
CSCD
北大核心
2023年第11期3327-3332,共6页
Application Research of Computers
基金
国家自然科学基金资助项目(71762017)
湖南省教育厅科学研究重点项目(19A077,20A081,21A0592)。
关键词
资源分配
云计算
动态定价
机制设计
resource allocation
cloud computing
dynamic pricing
mechanism design
