Research And Application Of Streaming Media Load Balancing Based On Nginx

Nowadays,with the rapid development of the network communication environment and the reduction of mobile tariffs,streaming media applications and entertainment have occupied a large proportion in daily activities in the context of China's 4G popularization and 5G development.As the mainstream and open source web server,Nginx provides support for streaming media server and load balancing functions.Analysis and research of Nginx can reduce server-side pressure,improve the delay and stability of streaming media playback,and enable users to have a better experience.Although the Nginx function is supported by modules,there are few examples of combining them.This thesis studies and builds a streaming server cluster suitable for streaming media load balancing algorithms to make Nginx performance and streaming media better.At present,Nginx provides a variety of conventional load balancing algorithms as built-in strategies for load balancers,but because conventional algorithms are limited to their own static,unreasonable weight configuration,and more emphasis on general conditions such as the number of requested connections,it is not suitable for use.Streaming media server equipment with a large load output;at the same time,the conventional algorithm can not satisfy the dynamic changes of the back-end server,and can not make effective judgments for the load situation or the instantaneous high concurrency.Therefore,this thesis designs a dynamic based on the existing conventional algorithms.The load balancing algorithm can dynamically feed back the server back-end performance metrics and meet the streaming server cluster requirements.This thesis proposes a dynamic feedback smoothing weighted round robin algorithm based on Nginx.This algorithm improves the load weight change process and optimizes the load performance of the backend server.The main research results of this thesis are as follows:1.Aiming at the problem that the conventional weighted polling algorithm can not dynamically feedback the load situation,this thesis dynamically calculates the load weighting information of the back-end feedback through the analytic hierarchy process and updates it.Simulation results show that the algorithm effectively improves the accuracy of server allocation.2.Nginx server lacks the smoothness of server allocation and the dynamic sensing ability in the face of high load.For this reason,the smoothing algorithm is used to make the server allocation more balanced and load migration function.The simulation experiment and system experiments prove that the algorithm is effective.Improve the stability and reliability of the streaming server.3.Based on the improved algorithm and Nginx,this thesis designs and implements a streaming media server cluster system suitable for airborne cabin entertainment services,and completes the system requirements analysis,overall architecture design,function module design,etc.The implementation of the system,the installation and configuration process,and the functional test and performance test of the system are introduced.