Research On Video Caching And Processing Algorithms In Edge Computing

With the rapid development of the Internet,caching technology is used more and more widely,from the original operating system and CPU to today's servers and edge nodes.With the emergence and popularity of concepts such as edge computing,caching has a richer application scene and landing mode.In general,caching hot content can reduce unnecessary overhead and increase efficiency.In the scenario of edge computing,taking advantage of the feature that edge nodes are close to users,caching popular videos in advance in edge nodes can shorten the time for users to acquire videos and greatly improve user experience.However,there can be many popular videos,individual videos can be encoded at different sharpness,and how to select videos and bit rates to cache in complex scenarios,how to balance user experience,limited storage and computing resources at edge nodes,and the cost of video providers(VPs)remains extremely challenging.In addition,in the case of multiple video providers,it is necessary to consider and study how the Network Service Provider(NSP)maintaining the edge node allocates storage space to each VP,improves space utilization,and maximizes its profit.Therefore,the main work of this paper is as follows:(1)Design method of multi-bit rate video cache in edge computing.Firstly,a parameterized system is established according to the popularity distribution of the video and the time delay of the user obtaining the video.Secondly,the cache benefits are quantified based on the utility function,so that modeling the video cache problem as the objective function is an optimization problem of the utility function.Then,this paper studies the scenes where the video contains only a single bit rate version and the scenes where it contains multiple bit rate versions,and proposes a video cache algorithm that is suitable for the two scenes respectively.Finally,the performance of the algorithm is evaluated by simulation experiments.The results show that the proposed algorithm has good performance and can provide a reasonable video caching scheme efficiently.(2)Design method of storage space allocation for joint video cache.In this paper,a parameterized system is established,and the distribution benefits are quantified based on the utility function,so that the spatial distribution problem can be transformed into an optimization problem with one constraint.After that,this paper tries to introduce the relevant methods in game theory to assist the solution,and puts forward the algorithm used to find the optimal solution and the equilibrium point of the game.The simulation results show that the proposed algorithm gives a reasonable space allocation scheme.