Research On Resource Allocation And Cache Management In Mobile Edge Computing

With the thriving of mobile communication technology and the popularity of mobile devices,the platform services have been expanded on mobile terminals.However,many services are restricted on mobile devices owing to complex computing or too high latency requirements,e.g.,autonomous driving services and mobile blockchain services.In order to solve the problem of limited resources of mobile devices,mobile edge computing can provide computing task offloading services,transfer computing tasks of application services to edge servers near the mobile device,and provide users with computing and storage resources according to effective resource allocation methods.At the same time,in order to complete computing tasks more efficiently,the edge computing network also provides a cache function to ensure that delay-sensitive and context-aware applications can be effectively performed in the edge computing network.The edge computing network also provides a caching function to ensure that delay-sensitive and context-aware applications can be effectively performed in the edge computing network.Based on the investigation and analysis of mobile edge computing,this paper investigates and analyzes the resource allocation method and cache management under edge-cloud collaboration,and studies the efficient resource allocation and cache mechanism in intensive computing offloading scenarios.The main contributions are as follows:(1)An allocation mechanism based on the combinatorial double auction is proposed for cloud-edge computing resource allocation.In the mechanism,the computing tasks in generating a new block and broadcasting throughout the whole mobile blockchain network to make consensus can be offloaded to C-ESPs.Resource allocation and pricing algorithms are proposed to determine winners and final prices,which satisfy the economic attributes of auction truthfulness,individual rationality,budget balance,and computation efficiency.And the proposed mechanism is simulated in various situations.The simulation results show that the proposed mechanism is effective with large-scale user group participation and outperforms in terms of resource utilization.(2)For the computing offloading of intensive real-time computing tasks with high latency sensitivity,a multi-content caching mechanism based on a three-tier edge computing network is proposed.Based on edge-cloud collaboration,a three-tier cache structure of edge-edge data center-cloud data center is used.And it can solve the problems of insufficient storage space in the traditional two-tier structure,long task calculation time and transmission time,which lead to task timeouts and user services that cannot be completed.This mechanism caches user requests for services and calculation results at the same time,abandons the current content popularity calculation method,comprehensively considers the factors that cause the task calculation time to be too high.In this paper,we redefine the service content popularity and task results in combination with the transmission distance and the value calculation strategy,which is used for content placement and cache replacement for different cache contents.The simulation results show that the caching mechanism can effectively improve the completion rate of the uninstall service and the cache hit rate within the time allowed by the user,thus proving that the mechanism can effectively reduce the delay of the uninstall service.