Design And Implementation Of Edge Cache Node Selection System In MEC

With the development of edge devices such as mobile Internet,Internet of things,artificial intelligence and hardware terminal,data flow presents explosive growth.However,the centralized big data computing model(such as cloud computing,etc.)can no longer efficiently process the traffic data of peripheral devices,such as smart phones,smart homes and other terminal devices.In order to alleviate the pressure of the core network,content distribution network(CDN)technology development,but still the problem of storage computing resources such as tension,and large time delay and high cost disadvantage,so the edge of mobile computing(MEC)-will continue to sink to the server are closer to the user side of the network edge become more attention from the academia and industry.Among them,edge cache is one of the important application scenarios of MEC,that is,edge cache node is deployed on the edge of mobile network to alleviate the pressure of backhaul link bandwidth,reduce the transmission of repetitive content and reduce network delay.But close to the user side facing deploying cache node number,cache space is limited,and the challenges of small coverage area,therefore in the area near the users to retrieve nodes are available,and within those cache node and execute effective replacement strategy is to select content edge caching technology need to solve the problems in the process of development,these problems are also the focus of this article research content.This paper designs and implements a simple simulation system for retrieving edge cache nodes and controlling the replacement of cache contents.To solve the problem of node retrieval,is proposed in this paper including graphics retrieval and space text retrieval two node retrieval scheme,including graphic retrieval scheme combines the genetic algorithm,and ray method to improve the efficiency of retrieval,the text retrieval scheme is based on the combination of text processing the language model to improve the accuracy and recall rate of the retrieved results.In order to improve the utilization of cache space and cache file hit ratio,this paper combines markov process with hierarchical storage strategy to achieve the effect of cache content prediction and the requirement of cache efficiency improvement.This paper first introduces the current research status of edge computing,spatial retrieval of edge nodes and text retrieval technology at home and abroad,and then analyzes,designs and implements the system.Among them,it focuses on the algorithm involved in the function realization,and finally tests the system to meet the relevant requirements of demand analysis.