Research And Optimization Of Service Platform For Merchants In Scenic Spots Based On Microservices

Merchant management in scenic spots is an important guarantee for the sustainable development of scenic spots.However,the mismatch of information between scenic spots and merchants makes it difficult for scenic spots to attract enough high-quality merchants.Many small merchants with mixed good and bad influences the tourist experience.Some scenic spots try to strengthen merchant management through information platforms.But most of them are developed and maintained through traditional software architecture and technology.There is room for further optimization at efficiency,performance,availability,and business scalability.This thesis designs and implements a service platform for merchants in scenic spots,mainly for scenic spots,merchants and tourists.It aims to provide merchants with one-stop services such as information inquiry,entry application and fee payment through information methods,which improves the management level of merchants and visitor experience.It contains six modules under the microservice architecture.Merchant management module provides services such as merchant information authentication and fee payment.Store management module provides services around store information and investment information,covering specific functions such as shop rental and publishing investment services.Interactive communication module provides visitors with evaluation feedback.Search statistics module combines with location search and information statistics.Besides,it helps tourists to find suitable merchants.It also provides statistics on tenant entry information.Authentication and third-party interfaces module provide third-party services such as user information authentication and SMS respectively.This thesis also improves the original system load balancing strategy.It adopts a load balancing strategy based on combination weighting to improve its performance and operating efficiency.Based on the original load indicators of the Ribbon load balancer,three node hardware indicators are regularly collected and introduced through the Linux command and microservice health detection.It contains CPU idle rate,memory idle rate and disk I/O idle rate.The weight value of the index is obtained by three methods.It contains combination weighting strategy based on the ideal scheme and analytic hierarchy process and entropy method.So that the relative weight of each load index can be got more accurately.Finally,the weight value and the data size of each load index are combined to get the weight size of each service instance,which guides the request to the node with lighter load.Compared with the original Ribbon strategy,experiment shows the average response time of this strategy is reduced by 5% to 10% when the amount of concurrency is large.That indicates the strategy optimizes the performance of each service and improves the user experience.The service platform for merchants in scenic spots based on microservices meets the expected functional requirements.It has good disaster tolerance and high concurrency performance.The test results verify the advantages and feasibility of the application of the microservice architecture in the system,so it has good application value.Combined with the built-in Eureka and the gateway of Zuul,Spring Cloud microservice framework is adopted to improve microservice management.At the same time,Spring Security OAuth2 framework is used to realize user authentication and permission control and improve system security.To further improve the system performance and reduce the coupling degree of various services,Rocket MQ is used to realize message decoupling and current limiting.At the database level,My SQL and Redis cache are used.Cuckoo filter is introduced to prevent cache penetration.At last,the functional and nonfunctional tests of the whole system are carried out.The service platform for merchants in scenic spots based on microservices meets the expected functional requirements.It has good disaster tolerance and high concurrency performance.The test results verify the advantages and feasibility of the application of the microservice architecture in the system,so it has good application value.