This documentation is preliminary and is not intended to be a complete manual. In order to understand the following, we recommend to read the M-tree paper at least. Please, refer to Bibliography.

The implementation of M-tree is built on top of the MESSIF library developed at Lab of Searching and Dialogue at the Faculty of Informatics, Masaryk University, Brno under the leadership of prof. Pavel Zezula.


The currently available implementation of M-tree tries to incorporate "the best" of extension available. The split strategy is based on building the minimum spanning tree defined in Slim-tree. It also supports the PM-tree extension based on using an independent set of pivots for additional pivot-filtering.

Java Documenation

Generated Java documentation from the latest source code:


List of features:

  • Range search algorithm
  • Incremental nearest neighbor (NN) search algorithm
  • kNN search algorithm - currently implemented by using the incremental NN algorithm. In this respect, it is not optimal and will be implemented efficiently in the future.
  • Insertion algorithm - insertion of one object based on multi-way insertion algorithm.
  • Bulk-loading support is not available yet!
  • Deletion is not supported yet!
Last modified 7 years ago Last modified on Mar 4, 2011, 12:30:22 PM