An Efficient Cache Conscious Multi-dimensional Index Structure

Abstract

Recently, to relieve the performance degradation caused by the bottleneck between CPU and main memory, cache conscious multi-dimensional index structures have been proposed. The ultimate goal of them is to reduce the space for entries so as to widen index trees, and minimize the number of cache misses. They can be classified into two approaches according to their space reduction methods. One approach is to compress minimum bounding regions (MBRs) by quantizing coordinate values to the fixed number of bits. The other approach is to store only the sides of MBRs that are different from their parents. In this paper, we investigate the existing multi-dimensional index structures for main memory database systems through experiments under the various work loads. Then, we propose a new index structure that exploits the properties of the both techniques. We implement existing multi-dimensional index structures and the proposed index structure, and perform various experiments to show that our approach outperforms others.