【摘要】 当数据立方查询条件不是合取范式时，一般是将它转化成为若干合取范式的并的形式(析取范式)。但如果各合取范式之间有交集，则交集部分的记录会被多次查询。为了解决这个问题，文章提出了一种数据立方查询条件优化策略，把查询条件转化为互不相交的立方块的并的形式。文章详细地讨论了数据立方中互不相交的立方块的划分方法，并给出了该优化策略的实现算法和性能分析。结果表明，当查询条件不是合取范式时，该优化策略明显提高了查询性能。更多还原

【Abstract】 When the constraint expression of a query is not a CNF (Conjunctive Normal Form),it is generally transformed into an equivalent DNF (disjunctive normal form)an union of CNFs . But if the intersections of CNFs are not null, the records in the intersections will be queried one more time. In order to resolve this problem, a query constraints optimal strategy for data cube query is discussed in this paper. The constraint expression is transformed into a union of cube chunks, which don’t intersect with each other in the optimal strategy. This paper is focused on partition of cube chunks and implementation and performance analysis of the strategy. And the result shows that this optimal strategy can improve query performance saliently if the query constraint expression is not a CNF.更多还原