Spatial data partitioning based on the clustering of minimum distance criterion

With the constantly expanding of geo-spatial data scale, the increasing complexity of spatial operation, traditional single GIS mode can no longer meet new requirements of the mass geo-spatial data operation. How to maximize parallel equipment computational ability is the hot research topic. The parallel task division and geo-spatial data partition are put on the strategy study as the precondition of GIS further performance. The paper is facing to the high performance parallel GIS operation demands and proposes a spatial data partitioning algorithm based on the minimum distance clustering, realizes load balance when partitioning spatial data. Designing a new way to fix the clustering centers based on k-means algorithm, the centers arranged according to the ascending x coordinate sort order and distributed even in the space. The experiment shows this algorithm has good results for spatial data partitioning while the clustering performance of spatial objects and load balance are taken into account.