| 摘要: |
| 数据压缩是提高多维数据仓库性能的重要途径,联机分析处理是数据仓库上的主要应用,Cube操作是联机分析处理中最常用的操作之一.压缩多维数据仓库上的Cube算法的研究是数据库界面临的具有挑战性的重要任务.近年来,人们在Cube算法方面开展了大量工作,但却很少涉及多维数据仓库和压缩多维数据仓库.到目前为止,只有一篇论文提出了一种压缩多维数据仓库上的Cube算法.在深入研究压缩数据仓库上的Cube算法的基础上,提出了产生优化Cube计算计划的启发式算法和3个压缩多维数据仓库上的Cube算法.所提出的Cube算法直 |
| 关键词: 数据仓库 压缩的数据仓库 OLAP(on line analysis processing) Cube |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金资助项目(69873014);国家重点基础研究发展规划资助项目(G1999032704) |
|
| Cube Algorithms for Very Large Compressed Data Warehouses |
|
GAO Hong,LI Jian zhong
|
| Abstract: |
| Data compression is an effective approach to improve the data wharehouses. On line analysis processing (OLAP) is the most important application on the data warehouses, and Cube is one of the most operators in OLAP. Thus, it is a big challenge to develop efficient algorithms for compressed data warehouses. Although many algorithms to compute Cube have been developed recently, there is little to date in the literatures about Cube algorithms for compressed data warehouse. To the authors' knowledge, there is only one paper that presented a Cube algorithm for compressed data warehouses with a special compression method called chunk-offset. A set of Cube algorithms for very large and compressed data warehouses are proposed in this paper. These algorithms operate directly on compressed datasets without the need of decompressing them first. They are applicable to a variety of data compression methods. The datail analysis of I/O and CPU cost are also given, and compared with the existed algorithms by experiment. The analytical and experimental results show that algorithms proposed in this paper are more efficient than other existed ones. |
| Key words: data warehouse compressed data warehouse OLAP （on line analysis processing） Cube |
