Abstract. In this paper, we present a fast algorithm to answer rangesum queries in OLAP data cubes. Our algorithm supports constanttime queries while maintaining sub-linear time update and using minimum space. Furthermore, we study the trade-off between query time and update time. The complexity for query is O(2d ) and for updates O((2 2 n)d ) on a data cube of nd elements, where is a trade-off parameter. Our algorithm improve over previous best known results.