We introduce the B-List Data structure as a practical and I/O efficient techniques for implementing the Chazelle-Guibas fractional cascading data structures. A nice feature of using B-Lists is that the initial search at the root of the query tree is itself implemented using B-trees. This is applied, in particular, to the problem of point location in monotone subdivisions. The use of monotone subdivisions for point location is more practical than triangulations because of improved space and time efficiency. For very large data sets, the space improvement can be critical.
Here is a demo using the map of Manhattan. A paper describing our approach may be found here.
This research is conducted by Yunyue Zhu and Chee Yap.