Abstract
Multiple Representation Databases are structures that link different data sets based on common geometry and/or semantics. They offer the possibility to realize an efficient updating process in case of changes, as only one data set has to be updated and the changes can be propagated to the linked one. There are some prerequisites for the realisation of such a functionality which are elaborated on in the paper. First of all, functional dependencies between the data sets have to be known or made explicit. Secondly, partitioning schemes of space have to be available in order to make sure that local changes can be restricted locally and do not spread over the whole data set.
Users
Please
log in to take part in the discussion (add own reviews or comments).