A Scalable Algorithm for Bichromatic Reverse Nearest Neighbor with Grids

Bichromatic reverse nearest neighbor (BRNN) has been extensively studied in spatial database literature. While previous algorithms for BRNN queries rely mainly on in-memory and do not guarantee the scalability. A straightforward approach is to determine the BRNN for all possible points that are not feasible since there are a large or infinite number of possible points. To the best of our knowledge, the fastest known method has exponential time complexity on the data size in parallel. Based on some interesting properties of the problem, we come up with a novel grid-based algorithm for Scalable Bichromatic Reverse Nearest Neighbor queries (SBRNN, for short) to answer large scale MaxBRNN queries. In addition, our approach is the first attempt for distributed BRNN queries based on the Scalable Grid Index and can be implemented on the cloud platform. Extensive experiments are conducted to show that SBRNN is efficient, scalable and outperforms previous techniques on both real and synthetic datasets on cloud environment.