I'm using HPE Vertica Place for processing large amounts of lat/lon points.
I need to find the nearest neighbor of a point in a different set of points.
Right now the queries I try to use join the two sets, then I use row_number ordered by ST_Distance between the two points and partitioned by the first point.
Obviously, this query takes a long time to run and is not scalable.
I know that vertica can index polygons for fast querying of point intersections, so the best solution I can think of myself right now is to construct a Voronoi diagram and load it as a set of polygons to my db.
Is there a better solution for this problem?
