Compressing multidimensional structures: a case study

The OnLine Analytical Processing (OLAP) operate on the information from the Data Warehouses, pre-calculating and processing all combinations of the group-by operator and materializing them in Multidimensional Structures or Data Cubes. It is a computational task to realize that needs time, space to store the data. Many studies presented various techniques oriented to the Data Cube compression like the Dwarf, Condensed Cube, BU-Condensed Cube, Min Cube, Prefix Cube and Quotient Cube. This paper presents a study of the difficulty to process the Data Cube and we implemented and applied the Prefix Cube and the Bottom-Up Condensed Cube comparing the performance with the traditional Data Cube processing. Good results were achieved reducing in 70% the traditional Data Cube in different types of data sets.