Improving the Performance of OLAP Queries Using Families of Statistics Trees

We present a novel approach to speeding up the evaluation of OLAP queries that return aggregates over dimensions containing hierarchies. Our approach is based on our previous version of CubiST (Cubing with Statistics Trees), which pre-computes and stores all possible aggregate views in the leaves of a statistics tree during a one-time scan of the data. However, it uses a single statistics tree to answer all possible OLAP queries. Our new version remedies this limitation by materializing a family of derived trees from the single statistics tree. Given an input query, our new query evaluation algorithm selects the smallest tree in the family which can provide the answer. Our experiments have shown drastic reductions in processing times compared with the original CubiST as well as existing ROLAP and MOLAP systems.