M-IVTT: An Index for Bitemporal Databases

In this paper we present an indexing structure for bitemporal databases which allows the user to tune the trade-off between query processing speed and index size. The structure is based on incremental trees indexing valid time ranges, which are pointed to by a tree indexing transaction time. Valid time trees are kept full at every so many transaction time points, and in between those, patching sets are kept. The valid time trees are B+-trees which efficiently index ranges via a mapping function. Analytic performance results are presented and discussed.