并行数据仓库的建模、OLAP界面的设计、数据提取和数据维护

【作者】 刘勇；

【导师】 李建中；

【作者基本信息】 黑龙江大学 ， 计算机应用， 2001， 硕士

【摘要】 数据仓库是信息技术的新焦点，它用来存储供查询和决策分析用的集成化和历史化数据。国内外已经对数据仓库开展了许多研究工作，企业界也纷纷推出自己的数据仓库产品。如INFORMIX公司的MetaCube和ORACLE公司的Discover及Express系列产品。但是，这些产品仍是采用关系数据库管理技术来实现的，大多数只是对原有的系统稍加改进。这些产品都具有很大的局限性。鉴于上述原因，我们研制了一个基于机群并行计算环境的并行数据仓库管理系统（PDWMS）。它采用了多维数据模型，支持数据仓库上的各种复杂操作和决策分析。 本文的工作重点放在了对PDWMS的建模、OLAP界面的设计与实现、数据提取和数据维护方面，提出了并行数据仓库的建模算法，一种新的OLAP界面设计方法，并行数据仓库的数据提取算法，并行数据仓库的通用数据维护算法。这些算法已经在PDWMS中实现，取得了良好效果。更多还原

【Abstract】 Data Warehouse is a new focus of information technology. It stores integrated and historical data which supports query process and decision- making analysis. A lot of research work has been done at home and abroad. Many corporations have developed their data warehouse products, such as the MetaCube of INFORMIX corporation, Discover and Express of ORACLE corporation. But, these products are based on the relational database management system and there are many restrictions in supporting decision- making analysis. So, A parallel data warehouse management system(PDWMS) based on parallel computing environment has been developed by our group. It has adopted a multi-dimensional data model, supporting all kinds of complicated operations and decision-making analysis. The focus of this paper is on how to create prototype, design OLAP interface, extract data and maintain data for parallel data warehouse (PDW). In this dissertation, we propose the following algorithms and methods: creating prototype algorithm for PDW, a kind of new OLAP interface designing method, extracting data from many data source algorithm and maintaining data algorithm for PDW. These algorithms and methods have been used in PDW. It has been shown that they are effective and efficient.更多还原