摘要
联机分析查询处理是一种涉及大量数据的即席复杂查询,它通常都包含分组聚集运算。分析了关系数据仓库星型模式存储结构和数据更新的特点,把实体关系看成分布式数据库中以内存排序缓冲区大小为分段条件的全局关系,对分组操作进行分布式聚集运算,给出了一种改进的MuSA算法,有效地提高了算法性能。
The OLAP queries are ad-hoc, complex queries and the process generally includes aggregate operation. This paper analyzes the characteristics of the star schema storage structure and the data update in the relational data warehouse, considers entity relation as the overall relation which is segmented by the size of memory sorting buffer in the distributed database, and combines the distributed aggregation operation with the sorting operation. A modified MuSA(sort-based aggregation with multi-table join) algorithm is proposed and the performance of the new algorithm can be improved dramatically.
出处
《计算机工程》
CAS
CSCD
北大核心
2005年第2期77-79,共3页
Computer Engineering
基金
国家科技部"973"前期专项基金项目"针刺与汉语认知脑功能成像的神经信息学研究"(2001CCA00700)
