Dynamic behavior of balanced NV-trees

Recently, some approximate high-dimensional indexing techniques have shown promising results by trading off quality guarantees for improved query performance. While the query performance of these methods has been well studied, however, the performance of index maintenance has not yet been reported in any detail. In this paper we focus on the dynamic behavior of the NV-tree, which is a disk-based approximate index for very large collections. The NV-tree has several configuration and implementation options that affect the performance of index maintenance. We report on an initial study of the effects of these options on the dynamic behavior of the NV-tree, and show that with appropriate implementation, significant performance improvements are observed.