Re: [postgis-users] Voronoi polygons for large dataset

2018-06-11 Thread Arnaud L.
Hi Nathan Le 07-06-18 à 17:15, Nathan Einstein a écrit : Unfortunately, it doesn’t seem like there’s any simple “divide-and-conquer”-type solution; because the Voronoi polygons on the edges will be irregularly shaped, I haven’t been able to come up with a clean way to combine the results from

Re: [postgis-users] Voronoi polygons for large dataset

2018-06-11 Thread Komяpa
Hi! Is the dataset publicly available? Can you file a ticket at https://trac.osgeo.org/postgis/newticket and get dataset attached to it? For divide-and-conquer kind of solution, you need to track bounds of "beach line" which is made of a lot of parabolic segments, and that is indeed not trivial i

Re: [postgis-users] Voronoi polygons for large dataset

2018-06-09 Thread Giuseppe Broccolo
Hi Nathan, 2018-06-07 16:15 GMT+01:00 Nathan Einstein : > Hi all, > > I’d like to create Voronoi polygons for a set of about 11 million points. > ST_VoronoiPolygons works fine for a subset of ~3 million of these points, > though the function’s processing speed seems to scale non-linearly in the >

[postgis-users] Voronoi polygons for large dataset

2018-06-07 Thread Nathan Einstein
Hi all, I’d like to create Voronoi polygons for a set of about 11 million points. ST_VoronoiPolygons works fine for a subset of ~3 million of these points, though the function’s processing speed seems to scale non-linearly in the number of points. I therefore haven’t been able to get it to work fo