Design And Implementation Of Goods Sales Analysis Platfoms Based On Hive

When faced with massive consumer behavior data,it is difficult for traditional relational database management systems to dig out and analyze the intrinsic links between consumer data.Hadoop distributed storage and parallel computing big data processing technology provides a feasible solution for the analysis of massive data.This paper presents an implementation scheme based on the Hive commodity sales analysis platform under the Hadoop platform.Discussing the process of building a multidimensional analysis platform based on Hive’s merchandise sales,managers can intuitively obtain multi-dimensional analysis information of merchandise sales to facilitate the business decision-making.The paper first describes in detail the analysis requirements of sales data from customers,products,time,sales channels,and sales regions,and designs corresponding system use case diagrams.Then,the paper gives the design process of the sales multi-dimensional analysis platform,designs the overall functional structure and database structure of the system,and gives the main database table content of the system.Hive is a data warehouse analysis tool based on Hadoop.It provides SQL-like query functions and converts Hive QL queries into Map Reduce programs for statistical analysis of massive data.The paper then elaborates the process of product sales analysis platform construction,realizes the main functions of the system,and provides key technologies for ETL automation implementation.It uses sqoop command scripts combined with Linux system timing commands to implement ETL automation,and business function data processing is all in Hive.HQL is used for processing,and different HQL statements are processed according to different business logics.Because HQL can be converted into map reduce statements,query analysis can be performed efficiently and in parallel.Use echarts 3.0 controls to provide users with a friendly web interface and intuitively obtain analysis results from charts.Finally,the paper tests the main functions of the system,designs test cases and analyzes the test results.The system achieves the expected goals.The paper uses Hadoop and Hive to build a sales analysis platform,which reduces system construction costs,improves fault tolerance,scalability,and the efficiency of processing data.The successful implementation of the sales analysisplatform enables managers to have a deep understanding of the company’s overall business status,avoid blindness,help the leaders make the right decisions,and increase the company’s sales and sales profits.