A Cost Optimal Strategy For Transcoding And Caching In Mobile Video Service

With the high-speed development of the Mobile Internet, mobile terminals represented by smartphones have become the main equipment of video playing. In order to satisfy the diversity of mobile terminals in the requirements of video file format and resolution ratio and fit the changing mobile network environment, the server-side should cache a large amount of video files in different formats and resolution ratios. The introduction of real-time transcoding will release the pressure of the cache. Meanwhile, the introduction of video segmentation mechanism will shorten the waiting time delay and promote the use ratio of cache space. Issa(Infrastructure As A Service) Cloud provides scalable computing and storage resources for real-time transcoding of a large number of videos. However, how to allocate the resources reasonably and realize optimal utilization of system resources to reduce the total system cost is still an urgent problem to solve.Aiming at reducing the total resource cost of real-time video transcoding system for mobile terminal video services, based on mathematical modeling and optimization techniques, optimizing the cache space capacity and the cache replacement policy as the main line, validated by optimization model, from two aspects of theoretical analysis and model verification, the optimization issue of videos’ transcoding and caching is studied in this paper.The main work of this paper includes the following three aspects:(1)Based on analyzing the requirements of mobile video services, it researches on the structure of real-time video transcoding system under cloud computing environment. Firstly, mathematical models for video transcoding relations, video contents management, unit transcoding and storage cost are established. Then, A mathematical model of the total system cost is established. Finally, it analyzes the impact of trade-off between large number of videos’ transcoding and caching on the total system cost.(2)In order to reduce the total system cost, a cost optimal strategy for video transcoding and caching is proposed. The calculation formula of the optimal cache capacity is established, and then the cache replacement strategy based on the maximum total values is proposed.(3)According to the modeling and optimization process, the optimization model ofvideo transcoding and caching in XPress-MP is established based on the mathematical model of the total system cost and the cost optimal strategy for video transcoding and caching. The optimization model is solved, verified and analyzed with simulated parameter values in XPress-MP environment, and the effectiveness of the cost optimal strategy is tested.Under the condition of simulated parameter values, the running results of the XPress-MP optimization model verify that the cost optimal strategy for video transcoding and caching can effectively reduce the total cost of the real-time video transcoding system.