Design And Implementation Of S2B2C New Retail Social E-commerce Platform

With the popularity of the new generation of mobile communication technology,the "Internet+ E-commerce" mode has developed rapidly,and online and offline sales have been gradually integrated,which has led to the concept of new retail.This paper finds that some traditional urban and rural grocery stores have the following disadvantages:high purchase cost,high risk of overstocking,low commodity turnover rate.However,most of these grocery stores are limited by the location,scale,cost and other aspects of the store,and can not be transformed into new retail by joining the new retail stores,the living space is gradually squeezed by Tmall store,Suning store and Family mart.Therefore,a platform to help these grocery stores transform into a new retail model and reverse the disadvantages of traditional retail has high social significance.From the technical level,the traditional single e-commerce architecture,the internal structure of the system is increasingly complex,it is difficult to support the long-term development of new retail business,the vertical application architecture is highly coupled and difficult to develop,expand and maintain,some distributed cluster architectures based on container technology are too large for the new retail mode of traditional grocery stores,with high development cost,which is not conducive to rapid prototyping and is difficult to adapt to the iterative agile development task with complex and changeable business requirements in the process of business model innovation.Based on this background,this paper designs and implements a new s2b2c retail social e-commerce platform based on Dubbo+ Zookeeper+Nginx distributed microservice architecture.Through the requirement analysis of the actual business process,the system business model and technical architecture design are completed,and the system is divided into six microservice modules by using the idea of microservice architecture.In addition,this paper realizes two independent systems:web supplier management system and data analysis center.At the same time,this paper implements distributed lock and distributed transaction,proposes an improved weighted minimum connection Nginx load balancing algorithm based on the real-time load state of the server,which further improves the platform's high concurrency and distributed cooperation ability.In addition,according to the demand of product recommendation,this paper implements a collaborative filtering recommendation algorithm,because the traditional algorithm does not consider the influence of time factor on the recommendation effect,this paper proposes an improved user score calculation method based on time weight,which further improves the recommendation effect.Finally,this paper completed the functional and non functional test of the system.In the non functional test,the improved Nginx load balancing algorithm and collaborative filtering recommendation algorithm are tested.Through system testing and actual online operation,it is proved that the system can meet the system performance requirements and the technical requirements of fast iterative development of business functions,and significant practical application value.