ABSTRACT
Use of a multidimensional access method in a spatiotemporal system offers the potential for improved performance in retrieving dimensional data. This chapter provides four possible methods of accessing spatiotemporal data, and examines how different partitioning approaches affect data accessibility. The four possible methods of accessing spatiotemporal data are grid file implementation, offset grid file implementation, R-tree implementation, and six-dimensional data space. The chapter describes mechanics, demonstration, and comments of grid file implementation, offset grid file implementation, R-tree implementation, and six-dimensional data space. The choice of R-trees is a simple one because it is the only irregular cell method in common use. The choice of the grid file and offset grid file is not straightfoward, however, because several alternatives are popular among researchers. The grid file offers strong procedures for treating fe-dimensional space, has a relatively flexible regular partitioning strategy, is clearly documented, and is elegantly simple. The most obvious alternative to the grid file would have been the quadtree.
