Data Warehouse Reporting Service System Or Enterprise Management Platform

With the development of big data era,the application of enterprise-level report plays an important role in corporate decision-making.The report data,regarded as one of summary data,provides basic data analysis support for enterprise management with its flexible presentation form and abundant multi-dimensional organization.The establish of efficient and available data warehouse model and report service system is the key of this paper.First,this paper analyzes the Web access characteristics of users in the enterprise-level report service system,and summarize the periodic access rules of users to monthly reports.In this paper,an improved cache replacement algorithm(GDSF-M),based on the report type and the current date weight function,is proposed to solve the problem of the low byte hit rate in the existing cache replacement algorithm.The proposed GDSF-M has been applied to the experiments to confirm the high hit rate and stability of the proposed method.It can effectively save the response time of the report service system and improve the system performance.Second,the disparate data sources are integrated and a global data warehouse model is established.As an important step of data warehouse construction,ETL task scheduling is the main time-consuming of data warehouse.The shortest time-consuming of achieving ETL task scheduling model is the vital research content in this paper.In the application scenario where the subject has a time constraint relationship,the existing genetic algorithm will increase the probability of occurrence of infeasible solutions,and result in more iterations and lower resource utilization.In this paper,an improved adaptive genetic algorithm IAGA-OF,using the weighted fitness function calculation method and the crossover and mutation operator with constraint relationship,is used to deal with the ETL task scheduling model.The experimental results demonstrate the validity of the model solving process,the convergence performance of the algorithm and the server load performance to achieve the purpose of the shortest ETL scheduling time.Finally,based on the proposed GDSF-M and the bottom layer of data warehouse model,the report service system is realized.In this system,the report calculation engine is used to build abstract metadata model.Meanwhile,the report calculation engine,which is compatible with fact tables and dimension tables,is able to support OLAP analysis and data queiry operations and achieve the self-service report module,personalized report module and role-based data access control module.The goal is to build a highly versatile and highly scalable reporting service system,which is also used to support dynamic porting.