Implementing Data warehouse Methodology Architecture: From Metadata Perspective

• 발행기관: 한국통상정보학회
• 저자명: Sang-youl Kim Tae-hun Kim
• 간행물 정보: 『통상정보연구』제7권 제1호, 55~74쪽, 전체 20쪽
• 주제분류: 사회과학 > 무역학
• 파일형태: PDF
• 발행일자: 2005.03.30

영문 초록

Recently, many enterprises have attempted to construct data warehousing systems for decision-support. Data warehouse is an intelligent store of data that can aggregate vast amounts of information. Building DW requires two important development issues:(i) DW for the decision making of business users and (ii) metadata within it. Most DW development methodologies have not considered metadata development; it is necessary to adopt a DW development methodology which develops a DW and its metadata simultaneously. Metadata is a key to success of data warehousing system and is critical for implementing DW. That is, metadata is crucial documentation for a data warehousing system where users should be empowered to meet their own information needs; users need to know what data exists, what it represents, where it is located, and how to access it. Furthermore, metadata is used for extracting data and managing DW. However, metadata has failed because its management has been segregated from the DW development process. Metadata must be integrated with data warehousing systems. Without metadata, the decision support of DW is under the control of technical users. Therefore, integrating data warehouse with its metadata offers a new opportunity to create a more adaptive information system. Therefore, this paper proposes a DW development methodology from a metadata perspective. The proposed methodology consists of five phases: preparatory, requirement analysis, data warehouse (informational database) development, metastore development, and maintenance. To demonstrate the practical usefulness of the methodology, one case is illustrated

목차

Ⅰ. Introduction

Ⅱ. Methodology Comparison and Past Studies

Ⅲ. Proposed Methodology

Ⅳ. Validation of proposed methodology

References

키워드

Data Warehouse Metadata Development Methodology Architecture