一种双样本多分形视图尺寸估计算法

A Multi-Fractal Double-Sample Algorithm for View Size Estimation

视图尺寸估计是数据仓库实化视图选择和分配预聚集视图存储空间的前提。本文提出了双样本多分形视图尺寸估计算法MDS;同时,为了研究MDS算法的有效性,本文把它与Cardenasn formula、SF 和FMS算法进行了比较。实验结果表明,MDS算法优于其它算法。

View size estimation is a fundamental step for selecting materialized views and allocating the storage space the pre compute aggregation requires in data warehousing. This paper presents an original Multi fractal Double Sample (MDS) algorithm for view size estimation. Meanwhile, to study the effectiveness of the algorithm, this paper compares its accuracy with the Cardenas n formula algorithm, the SF algorithm and the FMS algorithm. Experimental results show MDS is superior to the other algorithms.