Research On The Optimization Of Service Resource Allocation Based On Tullock Contest

The rapid advancement of information science and technology has brought profound changes to the development of society and economy.The improvemen-t of computing performance has spawned many novel Internet service industries(ISI).The new technology brings opportunities as well as enormous challenges,for example,the development of Internet services has brought huge energy consump-tion.It is necessary to investigate effective measures,strateg:ies,and algorithms for energy-saving actively.The new agreements and algorithms not only need to meet the traditional requirements of effectiveness and fault tolerance,but also need to conform to economic principles,such as cost,margin cost,and expected profits.Game theory and economic constraints can effectively lead to the for-mulation of new agreements and algorithms to achieve a win-win situation and achieve long-term economic benefits.This work aims to study the service resource allocation strategy in the Internet environment.According to the attribute clas-sification,resources of ISI can be divided into two categories:homogeneous and heterogeneous.Homogeneous resources refer to the resources belong to the same or similar structure,otherwise is heterogeneous,e.g.,cloud computing(homoge-neous cloud,heterogeneous cloud)and wireless network(homogeneous network,heterogeneous network).In ISI,service content is usually stored in a cloud data center,service providers complete user requests through cloud computing,and deliver services to users by wireless networks.This work exploits a framework of resource allocation for homogeneous and heterogeneous resource in the Inter-net service industry by using Tullock Contest.We extract the interests between service providers and customers,the competition relationship between service n-odes,and balance the conflicts of interests among the parties,which promotes all parties in the service to achieve a win-win situation.This work constructs an easy-to-use game theoretic model to balance the con-tradiction between economic growth and energy consumption in the emerging Internet services,the contradiction between the profit of service operators and demand of customers,to promote the smooth and continuous development of relevant,industries and improve economic efficiency.This paper mainly studies the equilibria of the N-players in Tullock Contests with the incomplete informa-tion configuration,which is an "NP-complete"problem.This work cornstructs an intelligent system tha.t,includes two parts:Monte Carlo methods and virtual repeated game,which allows "agents" to learn from itself and the best agents to get better st.rategies in the next round.In the end,a Nash equilibrium is reached.Based on this theoretical analysis,this work proposes a dynamic server re-source scheduling model in homogeneous cloud computing.The model focuses on optimizing the number of running servers in a cloud computing data center under different traffic.The service resources are scheduled according to the ac-tual traffic,the pricing model,and the server energy consumption.Under the premise of guaranteeing the quality of service,we allow for the appropriate loss of computat.ional performance to reduce energy consumption.In this way,a s-tate of equilibrium is achieved.Further,we ext.end the homogeneous model to the heterogeneous model,a power rationing strategy for small cells is proposed in a heterogeneous network.The conflict between Quality-of-Service(QoS)and energy consumption is converted to a tradeoff between profits and costs.The model consists of two parts.Firstly,a dynamic pricing function is used,which is related to the power rationing strategy and QoS of small cells.Secondly,the Nash equilibrium of power configuration is obtained by applying the Monte Carlo Methods and virtual repeated game to reduce the energy consumption and im-prove t.he profits of service providers.Since the energy consumption and revenue of the small cells are balanced,this ensures that the profits of the HetNets system can be maximized.The wireless network is t.he main bearer of data transmission in the ISI.users mostly use mobile smart devices to access various applications in the cloud through wireless networks.This a joint,model of wireless net.work and cloud computing covers almost all current Internet services.The results show the feasibility of the scheme and can be used as a tool for the tradeoff between profits and costs in other application scenarios.