Resource Allocation And Cache Service Framework In Dynamic Edge Network

Mobile Edge Computing(MEC),as a prospective computing paradigm,can significantly enhance computation capability and save energy for Smart Mobile Device(SMD)by offloading computation-intensive tasks from resource-constrained SMDs onto the Edge Cloud(EC).Compared to the central cloud,the edge cloud can provide services to nearby SMDs with lower latency.However,the SMD may be mobile,and the resources of the edge network are limited to multi-user.Therefore,how to obtain an optimal offloading policy under the constraints of mobility and limited resource remains a challenging issue.Edge Cache Service(ECS),as a prospective content distribution paradigm,can significantly reduce the data transmission latency and improve the Quality of Service(Qo S)of digital content providers by offloading content data to edge servers in the network.Compared to centralized content service,ECS can provide digital content from nearby edge servers via a high-speed wireless network with fewer hops.However,the edge cloud and network service provider will not provide free services,we need to design a reasonable mechanism to encourage edge devices to share their resources and provide computing service and caching service.Many of the existing research work does not consider the characteristics of edge network,such as dynamic and limited resources.Therefore,it is of great practical significance and application value to carry out the research of MEC resource allocation mechanism and related framework according to the characteristics of edge network.Combining the latest research works in MEC,this paper mainly studies the multi-user task offloading and resource allocation mechanism,resource dynamic pricing mechanism,and cache service framework in edge network.The main work and contributions of this paper include the following three aspects:1.This paper provides an optimal multi-user task offloading policy to minimize the total application execution cost of SMDs in the MEC system.First,it formulates the problem of SMD computation offloading and edge cloud resource allocation into the total execution cost minimization problem of SMDs.This paper measures the execution cost in terms of application execution time or energy consumption,under the constraints of limited connection time between SMDs and edge clouds,application completion deadline,and limited resources of edge clouds.Second,the total execution cost minimization problem is decomposed into two sub-problems.The first sub-problem is how to obtain an optimal offloading policy of SMDs.The second sub-problem is how to achieve the optimal edge cloud resource allocation policy.The maximizing SMD offloading payoff algorithm based on the game model is given to solve the first sub-problem.Furthermore,the multi-user payoff competition algorithm is provided for solving the second sub-problem.Finally,experimental results show that compared to existing algorithms,algorithms in this paper can significantly save the total energy consumption and reduce the total application execution time of SMDs in the edge network.2.This paper studies the MEC resource pricing mechanism.An effective pricing mechanism to jointly optimize both radio resource prices and computing resource prices in the MEC system by balancing the resource supply and resource demand in the resource market of the MEC system is provided in this paper.This paper builds a computing and network resource trading model by considering the limitation of the maximum budget and supplier sellable resources in the MEC system.The trading model allows the SMDs to customize budget allocation strategies based on their preferences to speed up the task execution or reduce the task execution energy consumption.The MEC system utility maximization problem is formulated under the constraints of resource limitation and task execution deadline.This paper provides the solution for this optimization problem to jointly adjust the network resource pricing,the computing resource pricing,and the budget allocation to maximize the system utility of the MEC system.This paper through the concept of portfolio investment in microeconomic theory to obtain optimal budget allocation strategy via solving the investment return maximization problem for the SMDs,which can significantly reduce the budget allocation strategy search cost.The equilibrium price finding algorithm based on the max-flow/min-cut theorem is provided to adjust the network resource prices and computation resource prices to achieve the market equilibrium by considering resource supply and resource demand in the MEC system.Simulation results show that the algorithms in this paper can obtain the budget allocation strategy within fewer iterations compared to state-of-the-art methods,and can maximize the system utility by finding the equilibrium price.3.This paper designs a decentralized edge caching service framework based on blockchain technology to enable trusted transactions of cached resources and trusted transactions of data content in the edge network.The framework allows content providers to offload their content data to edge devices for enhancing Qo S.This paper designs a cache order matching mechanism based on futures trading theory for matching cache resource transactions between content providers and edge devices,which significantly improves the utilization of cache resources.In addition,this paper provides a content trading mechanism with data integrity verification to support data sharing among edge devices.The designed transaction contract management mechanism can incentivize edge devices to complete their contracts and can ensure the fairness and validity of transactions in the edge cache service system.