The Optimization And Application Of Storage And Query Strategy Of MongoDB

The rapid development of the information age has invisiblely shortened the distance between people,making the frequency of interpersonal exchanges increase.At the same time,the total amount of data generated in the network exchanges is also growing and the types are also more diverse.Therefore,how to effectively manage and utilize these massive social network data is very important.With the rise and widespread use of social network services,social applications such as microblog have gradually become the media of online public opinion,including the launching points and dissemination platforms of various rumors.For college administrators,it is also necessary to face the problems of student microblogging users in the manufacture and dissemination of public opinion.Due to the large base of student users and their close association with each other,it is very easy to generate group retransmission incidents.Therefore,the development of independent school microblog public sentiment system is of great significance.Taking into account the current micro-blog data is moving toward a huge data size,data type diversified direction of development,the existing school microblog public sentiment system if you continue to use the traditional relational database will be many problems.Therefore,this thesis will introduce non-relational database on behalf of MongoDB to try to replace non-relational database MySQL as the underlying storage database for school microblog public sentiment system.This thesis mainly completed the following tasks:(1)Apply MongoDB to microblog public sentiment system to replace the MySQL to solve the problem of huge amounts of data of data management,in view of the microblog public sentiment system database storage requirements,design of the MongoDB database storage solutions in detail;(2)Based on the data storage model,the design method of embedded document mode is proposed to solve complex queries;(3)According to the auto-sharding mechanism of MongoDB,the selection mechanism of the shard key is proposed to solve the problem of uneven distribution of data between slices;(4)Based on the analysis of the MongoDB query optimization mechanism,the method of the Where-Limit improved query is proposed to resolve the inherent problems of the built-in skip-limit model;(5)Based on the deficiency of the built-in load balancing algorithm,a data equalization algorithm based on node load difference is proposed to solve the problem of unbalanced read and write load.The experiment proves that the application of MongoDB in the microblog public sentiment system after the improvement of storage and query strategy has achieved better results,and the performance of the system has been improved greatly.