摘要
随着多维数据分析在各领域的广泛应用,基于数据立方体的计算方法受到大量研究者的关注。分析了影响数据立方体计算的各种因素,其中包括数据存储空间、查询处理效率和数据立方体的维护消耗,并且阐述了数据立方体的物化策略。分别从冰山立方体、紧凑数据立方体、高维数据立方体、近似计算、流式数据立方体等几个方面综述了国内外现有的计算方法,分析了各种方法的特点以及适用范围。
With the wide application of multidimensional data analysis in various fields, data cube computation attracts more and more attentions of researchers. We analyzed the influencing factors of computation which include the size of storage space, the query processing efficiency, and the cost of maintenance, and discuss the strategy of data cube materialization. The existing approachs of data cube computation were reviewed from the aspects of iceberg cube, compressed cube, high dimension data cube,approximate computation, stream cube and so on, the property and suitable area of the approaches were discussed in detail.
出处
《计算机科学》
CSCD
北大核心
2008年第10期1-5,32,共6页
Computer Science
基金
国家自然科学基金(70771110)
