The modelling of high dimensional data sets pose a particular challenge both for approximation and the parallel implementation. In this project we investigate the parallelization of sparse grid methods.