Font Size: a A A

The Research And Application Of Data Warehouse Modeling In Telecom

Posted on:2007-03-26Degree:MasterType:Thesis
Country:ChinaCandidate:C YuFull Text:PDF
GTID:2178360185967687Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
The data warehouse is defined as a subject-oriented, integrated, non-volatile and time variant collection of data in support of management's decision. The telecom company is facing analyzing decision problems. The data warehouse can be used to solve these problems.The paper focus on enhancing the ETL efficiency by mending the data warehouse model. The data warehouse model is divided into two parts. One is center data warehouse model, the other is data mart warehouse model. So the subject designing is divided into center warehouse subject designing and data mart subject designing. The center warehouse is designed by normal form and the mart warehouse is designed by normal form and star model. The center warehouse is designed according to the enterprise model. The mart warehouse is created according to requirements. The mart warehouse is divided into base tables and fact tables. The fact tables is based on the base tables. The base tables are the summaries and joining of the center warehouse data. Then the work of ETL is divided into pre-ETL and post-ETL. The main work of pre-ETL is to map sourcing data into center warehouse data. The main work of post-ETL is to map center warehouse into data mart. The person working in pre-ETL needs to be familiar with sourcing system and center warehouse model. The person working in post-ETL needs to be familiar with center warehouse model and the requirements. At the same time the checking of data quality is done easily. And the problem is easily found.The application of the methodology proposed in this paper was discussed, based on an actural project. The subject domain of center warehouse is divided into seven subjects. The subject domain of data mart warehouse is divided into several parts...
Keywords/Search Tags:data warehouse, data mart, information modeling, ETL, telecom
PDF Full Text Request
Related items