P-CBF: A parallel cell-based filtering scheme using a horizontal partitioning technique

To efficiently retrieve high-dimensional data in data warehousing and multimedia database applications, many high-dimensional index structures have been proposed, but they suffer from the so called 'dimensional curse' problem, i.e., the retrieval performance becomes increasingly degraded as the dimensionality is increased. To solve this problem, the cell-based filtering (CBF) scheme has been proposed, but it shows a linear decrease in performance as the dimensionality is increased. In this paper, we propose a parallel CBF scheme using a horizontal partitioning technique, which is called P-CBF, so as to cope with the linear decrease in retrieval performance. To achieve it, we construct our P-CBF scheme under an SN(Shared Nothing) cluster-based parallel architecture. In addition, we present data insertion, range query processing and k-NN query processing algorithms which are suitable for the SN architecture. Finally, we show that our P-CBF scheme achieves good retrieval performance in proportion to the number of servers in the SN architecture and that it outperforms a parallel version of the VA-File when the dimensionality is over 10.