Design And Implementation Of Big Data Visualization System Based On IPTV

With the gradual popularization of digital TV and the continuous enrichment of terminal equipment,more and more users choose IPTV as their home Internet TV.IPTV provides users with diversified services such as live broadcast,on-demand,games,and conferences.Hundreds of millions of users have generated huge amounts of traffic data during the interaction with IPTV.To help operators understand user experience,monitor network anomalies,optimize CDN node deployment,and visually analyze traffic data.In the era of big data,data visualization has become an important part of data analysis.Direct visualization has the problems that the visualization graphics are too dense,the visualization objects cover each other,the GPU is overloaded,and the rendering is stuck.This paper designs and implements a big data visualization system based on IPTV.The big data architecture and trajectory clustering model are used to cluster the original data,and the processed data set is visually displayed.The main work is as follows:1.Aiming at the problems that the image is too dense,the object features are hidden,and the rendering is stuttered due to the large amount of data,a trajectory clustering based on reference center(TC-RC)is proposed.This algorithm improves the similarity formula between trajectories of the trajectory clustering algorithm and introduces the concept of reference center,which makes the clustering results more representative of the overall movement trend and effectively reduces the time complexity of the density clustering algorithm.2.The business logic of the visualization system based on big data is relatively complex,the project scale is large,there are problems with slow page rendering speed,front-end and back-end logic over-coupling,and poor scalability.For the above problems,the paper proposes a front-end microservice based on the Node middle layer Architecture.The architecture utilizes the logical processing capabilities of Node's middle layer to perform server-side rendering and request proxying,which improves page rendering speed.With the help of the single-SPA framework,front-end microservices are built to enhance the independence of modules and the reusability of components,Significantly improved efficiency.3.Designed and implemented a big data visualization system based on IPTV,used Hadoop distributed computing platform to complete the storage,calculation and analysis of massive data,and visualized the processed data set through B / S.