EXTERNAL BALANCED REGULAR (x-BR) TREES: NEW STRUCTURES FOR VERY LARGE SPATIAL DATABASES

The External Balanced Regular (x-BR) Trees constitute a family of new secondary memory structures which are suitable for storing and indexing multi-dimensional points and line segments. In 2 dimensions, the resulting structure is an External Balanced Quadtree, in 3 dimensions an External Balances Octtree, and in higher dimensions an External Balanced Hyper-quadtree. The main characteristic of all these structures is that they subdivide space (in an hierarchical and regular fashion) into disjoint regions. These spatial access methods are fully dynamic, while insertions are not complicated to program and aaect only one path in the tree. Moreover, x-BR trees are variable resolution structures. That is, the number of space subdivisions is not predeened, making these structures suitable for very large amounts of data. Due to the balanced nature of these structures and the disjoint-ness of the resulting regions, searches and other queries in these trees are processed very eeciently.