Modeling Short-form Video Transfer in Information Centric Network
Han Xu, Haozhe Wang, Jia Wen Hu, Zhengxin Yu · 2021
Over the past few years, short-form video has been gaining unprecedented popularity around the world. The explosive growth of user-generated content (UGC) puts tremendous pressure on the current communication networks. To meet the high bandwidth and low latency requirements of short-form video. Information Centric Networking (ICN), a promising future Internet paradigm, has been attracting ever-increasing attention from academia and industry. In-network caching and pending interest table (PIT) are two essential features of ICN that are designed to not only handle bulk data dissemination and retrieval but also to reduce the bandwidth consumption. The traffic generated by short-form video applications, which takes a substantial amount of the mobile bandwidth, has been observed to have the bursty characteristic. To improve the quality of short-form video services, it is important to have an analytical model that can accurately characterize the content transfer in ICN under different forwarding strategies and bursty traffic conditions. In this paper, we exploit the queueing network theory to develop a new analytical model for content transfer in ICN under bursty content requests and derive the mathematical expressions for calculating cache and PIT miss rate. The accuracy of our analytical model is validated by comparing the analytical results with those obtained from simulation experiments. We also use the model to investigate the content delivery time under various traffic conditions and forwarding strategies.