We have proposed an incremental maintenance scheme of data cubes employing extendible multidimensional array model. Such an array enables incremental cube maintenance without relocating any data dumped at an earlier time, while computing the data cube efficiently by utilizing the fast random accessing capability of arrays. But in practice, most multidimensional arrays for data cube are large but sparse. In this paper, we describe a data compression scheme for our proposed cube maintenance method, and demonstrate the physical refreshing algorithm working on the data structure thus compressed.