Dynamic Service Caching In Mobile Edge Networks

In recent years,with the rapid development of technologies such as the Internet of Things and the mobile Internet,various data services and applications have emerged.The emergence of data services and applications has greatly facilitated people's daily lives,but at the same time,it has brought enormous challenges to the existing network structure.As emerging data services and applications become more demanding for computing power and response latency,traditional mobile communication network architectures are in urgent need of new changes.Mobile Edge Computing was therefore proposed.Mobile edge computing provides computing power at the edge of a traditional mobile communication network by configuring a MEC server on the edge node,meeting the low latency requirements of emerging applications by the closer physical transmission distance between the edge node and the end-user.The service cache problem is a key issue in mobile application computing.Since various resources(such as bandwidth resources,computing resources,etc.)that each network edge server can provide to users are fixed and limited,only a few kinds of service requests can be provided and run by edge nodes at the same time.A reasonable mobile edge caching strategy can effectively improve the performance of the entire network.At the same time,due to the dense deployment characteristics of the next-generation network(5G network),the caching strategies of neighboring base stations are coupled in time and space,and the performance of the MEC network can be improved by coordinating the caching strategies by multiple base stations.For the dynamic service cache problem of mobile edge computing network,this paper establishes a reasonable caching strategy to enable the end user's service request to be localized as much as possible,so that the traffic load uploaded to the remote cloud computing center is minimized,thereby improving the whole Mobile edge computing network performance.This problem includes multiple difficulties such as federated service caching,end-user selection,and joint allocation of computing/communication resources.For the dynamic service caching problem of mobile edge computing without delay constraints,this paper first proposes a heuristic centralized and distributed service caching algorithm.Then,according to the dynamics of the mobile network,a context-based centralized and distributed dynamic service caching algorithm is proposed by adding aprediction mechanism,so that the caching strategy can make corresponding adjustments according to the user's changes.Experimental results show that the algorithm can reduce the service switching cost in the network while effectively reducing the transmission load transmitted to the remote center.Aiming at the dynamic service caching problem of the mobile edge under delay constraint,by adding low delay constraint to the edge cache problem of the aforementioned moving edge,the improved centralized and distributed algorithm under delay constraint is proposed for the targeted improvement.The strategy can effectively achieve the optimization goal under the requirement of low latency.The simulation results show that our algorithm can effectively reduce the transmission load transmitted to the remote center and improve the performance of the entire network.