Research On Performance Optimization Of Network Service For Health Data Management

As people pay more attention to health,the combination of the Internet and the physical examination industry has brought new opportunities to many enterprises.Physical examination industry achieves health information by the unique advantages of the Internet.Users can understand the health data in real time,and can compare historical data.However,the existing network service for health data management can not meet the growing needs of users,so this paper researches on performance optimization of network service for health data management to better serve people's health.Firstly,this paper introduces the application background and technical background,the concept of cluster,classification of cluster and system architecture.The classification,implementation and common algorithm of load balancing are expounded in detail.Secondly,this paper introduces the function and architecture of the health data management system,analyzes the performance bottleneck of the existing architecture,and proposes a new architecture of load balancing cluster network service for health data management.The application layer,load balance layer and database layer optimization scheme are introduced briefly.After analyzing the shortage of existing load balancing algorithms,this paper proposes a dynamic feedback load balancing algorithm based on weighted least connection.This algorithm takes into account the load index of the server and reflects the influence of each load index on the integrated load by giving different weight for each load index,and gives the formula and the adjustment method of server weight.Content analysis module,load balance module,state information collection module and performance data receiving module are analyzed in detail.Finally,we achieve a new dynamic feedback load balancing algorithm on Nginx.At the data service layer,the read/write control optimization is used to separate the health data.Using the Memcached cache database to cache high-frequency data,and through the Nginx server consistency Hash algorithm to achieve distributed cache database.This paper compares and analyzes the different architecture schemes and different load balancing algorithms,by using the testing tool.It proves the validity of the proposed load balancing cluster architecture.The new architecture has a significant effect on improving concurrent traffic and reducing response time.At the same time,it is proved that the dynamic feedback load balancing algorithm based on weighted least connection can deal with the change of load more effectively and make full use of system resources,by collecting load information in real time.