File Cache Value-Aware Cache Strategy In Edge Network

By offloading caching from the cloud to edge,edge caching technology represents a promising technology for improving Internet performance to meet the requirements of Internet latency-sensitive applications.An efficient cache strategy on edge caching devices is a key factor in ensuring cache hit ratios.The existing cache replacement policies(such as FIFO,LRU,LFU,and RR)ignore the influencing factors of file size and are likely to result in cache pollution problems.To improve cache space utilization and reduce user request delay,this paper proposes a novel method to determine the value of file cache based on the edge cache system model.A file cache value-aware cache replacement(FCVACR)algorithm is developed based on file cache value.The main work and research contents of this paper are summarized as follows:1.This paper presents a model of the network edge caching system.It explores the impact of the user request behavior,file popularity,and file size on the edge caching system and user request latency.In order to improve the cache hit rate and reduce the average user request delay,the problem of file cache decision and cache replacement in the edge cache system model is studied.2.Based on the characteristics of file size,the network edge cache decision strategy is given in this paper.The optimal cache threshold in the base station is obtained through a large number of simulation experiments.Under the optimal cache threshold,by caching some or all of the file to reduce the user's wait latency.Experimental results show that the proposed cache size decision-based cache decision strategy can improve the cache hit rate and reduce the average request latency of users.3.This paper studies the cache replacement strategy with the goal of minimizing average latency and maximizing cache hit rate by introducing file cache value.A cache replacement algorithm based on file cache valueaware is proposed.This paper sets up an experimental platform to demonstrate the effectiveness of the proposed method.The experimental results show that the proposed FCVACR algorithm is reduced by 14%,13.9%,12%,and 15.6% compared to the average delay of the FIFO,LRU,LFU,and RR algorithms,respectively.It can be seen that the FCVACR algorithm proposed in this paper is able to meet the user's low latency requirements for delay-sensitive network applications.