Design And Implementation Of A Distributed Shopping Platform For High Concurrency

The rapid development of the Internet has also led to the development of e-commerce at home and abroad.More and more people choose to implement commodity trading through e-commerce platforms.The increasing volume of online transactions has undoubtedly brought new challenges to the e-commerce platform architecture.The traditional SSM framework-based centralized architecture can not meet the system throughput and concurrency requirements.Under the access of large amounts of data,how to dealing with complex business logic and sudden peaks in access and ensuring that the e-commerce platform can provide users with reliable and stable services becomes a technical problem that need to be solved.In order to solve the above problems,this paper introduces a high-performance distributed service framework Dubbo which builds a new e-commerce platform architecture by integrating it with the SSM framework.Thus,a distributed shopping platform that supports high concurrency,load balancing,and scalability is realized.The main research contents of this paper are as follows:(1)Using the distributed service framework Dubbo which integrates Spring,SpringMVC,and Mybatis frameworks as the overall architecture of the shopping platform and using Zookeeper,Redis,Solr,ActiveMQ,FastDFS,Nginx and other technologies to complete the design and development of a distributed shopping platform that supports high load and high concurrency.(2)In order to improve the concurrency performance of the system,this paper uses Nginx as the reverse proxy to implement the load balancing of the Web,and based on that it adopts a load balancing algorithm optimize the performance of Nginx's load balancing which can dynamically adjust the weight.and uses the entropy method to the weighting coefficients of the four core elements in the algorithm thereby further improving the platform load balancing capability and concurrency performance.(3)Based on the above technical architecture and system,develop a full functioning and facing high concurrency shopping platform with a set of content management system,mall portal,search module,shopping cart module,order module,commodity details display module and user module.(4)Test the function and performance of the platform,compare and analyze the improvement effect of the platform's concurrency performance and throughput by the load balancing strategy selected by this paper.The test results show that the dynamic load balancing strategy used in this paper improves the platform's concurrent performance by about 17% compared to Nginx's built-in static load balancing strategy.Finally,summarize the work of this paper and analyze the shortcomings of the system and the direction that should be improved later.