The Research And Implementation Of High Performance And High Concurrent Network Application Architecture Based On Node.js

With the popularity of Internet and the advent of the era of big data,the daily traffic of mainstream network applications has reached trillion level.In the face of massive user concurrent access,how to ensure the high-quality and smooth experience has become one of the difficulties to be solved in the network application.In order to solve the problem of high concurrency,network applications need to enhance and improve the overall design and architecture from inside to outside,and improve the concurrency capabilities in all aspects.Not only the structure should be strengthened,but also the key algorithm should be optimized.And the network applications could also use third party technologies to improve concurrent capabilities.Node.js technology adopts an asynchronous I/O model based on event driven.Compared with the multilinear programming model,Node.js technology is more suitable for applications with high concurrency,big data and real-time response.In this paper,from the view of solving the high concurrency problems of network applications,the main research work is to design a high performance and high concurrency network architecture,research the load balancing algorithm and improve the load balancing algorithm in the network application architecture.The main work completed in the paper is as follows:(1)The article analyzes the commonly used network application framework types,summarizes the overall framework of the architecture and design principles of the internal structure.Aiming at the problem of high performance and high concurrency,a high performance and high concurrent network application framework is designed based on Node.js technology.The modules of the design scheme are the load balancing layer,core processing layer and data storage layer.Load balancing layer uses Nginx technology to apply load balancing algorithm.The core processing layer is based on the Node.js technology to complete the network application business logic processing.the memory cache service in the core processing layer adopts the Redis technology.The data storage layer uses the non-relational database Mongo DB to realize the data storage.(2)The article makes an in-depth analysis of load balancing algorithm which commonly used in research and the weighted polling algorithm,and improved weighted polling algorithm based on Nginx,which adds parameters such as CPU,memory,disk I/O rate,network bandwidth and response time to the process of weight calculation.Compared with the weighted polling algorithm,the improved load balancing algorithm is proved to be effective by testing and analyzing.(3)Based on the high-performance and high-concurrency network architecture and the improved load balancing algorithm,the paper realizes a blog application and displays blog functions.