Which Data Warehouse Architecture is Best?

被引：7

作者：

Ariyachandra, Thilini ^{[1
]}

Watson, Hugh J. ^{[2
]}

机构：

[1] Xavier Univ, Williams Coll Business, MIS, Cincinnati, OH 45207 USA

[2] Univ Georgia, Terry Coll Business, MIS, Athens, GA 30602 USA

来源：

COMMUNICATIONS OF THE ACM | 2008年 / 51卷 / 10期

关键词：

(Edited Abstract);

D O I：

10.1145/1400181.1400213

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Some of the significant factors that need to be considered before choosing data warehouse architecture, are discussed. The independent data marts provide a repository of decision support data for a specific application or business unit. The bus architecture has data marts to support orders, deliveries, or customer call for any business process. The hub and spoke architecture provide enterprise-level analysis of data requirement to create maintainable infrastructure. This architecture maintains atomic level data in the warehouse in normal form and create dependent data marts. The centralized architecture is similar to hub and spoke without any dependent data marts and contains atomic level data, summarized data, and logical dimensional views of the data. The federated architecture can be used for fragmented decision support data environment.

引用

页码：146 / 147

页数：2