The fields on which to partition a dataset. The partitions will be applied hierarchically based on the order of this list.