Cooperative Caching At The Edge Based On Deep Reinforcement Learning

The unprecedented growth of data traffic at the edge brings unique challenges for network bandwidth and server resources to meet the diverse Qo E(Quality of Experience).Caching becomes a promising way to alleviate these issues by storing a subset of data at the network edge,for which caching policy becomes critical.To this end,various caching schemes have been put forward,however,these schemes are either not intelligent lacking the ability of self-learning and self-decision-making,or inefficient with low data hit rate.Based on these observations,in this paper,we propose a novel Intelligent Caching framework at the Edge,named ICE,via deep reinforcement learning(DRL)to capture certain valued information of the requested data.Notably,in our approach,the popularity of the data to be cached is fully considered.A Markov decision model is further developed to determine whether the data should be cached.Moreover,in order to further improve the caching efficiency with higher data hit rate and higher reduction in transmission time,we further proposed a distributed computing-based multi-node cooperative caching scheme named DCCC.Specifically,each caching node in DCCC takes advantage of the DRLbased ICE scheme to optimize the hit rate,and then DCCC leverages a novel distributed computing-based method to merge the ICE's caching results of different caching nodes,based on which DCCC eventually manages the resource distribution aiming to improve the overall caching space utilization.Comprehensive experiments show that the singlenode ICE scheme greatly improves the cache hit rate and contents exchanging time in comparison with both DRL-based and legacy approaches,and our multi-node cooperative caching scheme DCCC further significantly improves the overall utilization of caching space.