摘要
多维数组进行存储通常是将其线性化为一维数组的方式进行存放,这种方法不利于数据的多维分析。首先,采用分块存储方法,将数据立方体划分为小的立方体为基本单位进行存储,然后,为每一个多维数据立方体创建一个数据文件,将划分后得到的有效数据块依次存放在数据文件的数据域中,在文件结束部分创建数据块的索引,即数据块在文件中的起始位置。
The multi-dimensional array is usually made into a one-dimensional array for storage by linearization, which is not fit for the multi-dimensional analysis of data. This paper adopts a new method of block storage, which divides the data cube into small blocks as the basic units for storage. Then it creates a database file for every multidimensional data cube and stores the valid data get by division into the data field of database file successively. Finally it creates the index of data blocks at the end of the file, namely the starting position of data blocks in the file.
出处
《微型电脑应用》
2015年第4期55-57,共3页
Microcomputer Applications
关键词
数据仓库
数据立方体
多维数据存储
Data Warehouse
Data Cube
Multidimensional Data Storage