I'm coming from a heavy Esri background and I'm trying to figure out how to create a multipoint feature in a PostGIS database that allows me to set a unique attribute for individual points in a multipoint feature. I've got data that is basically a huge number of XYZ points with an associated reading value (float) that are all grouped. I'd rather not create a database table with individual points as it will quickly baloon into the hundreds of millions of points. I've run databases with around 100 million rows before and it starts to get a little less performant past around 80 million rows in my experience.

I can always toss more hardware at it or implement table partitioning but I'd rather just use a multipoint and never run into issues to begin with.

  • What's your question? There are limitations to spatially diverse multipoints, but within reason, they will work. Still, this seems more like a discussion entrypoint, and GIS SE uses a "Focused question / Best answer" model. See the Tour for details. – Vince Jan 6 at 1:33

PostGIS geometry supports coordinates with up to 4 components, X Y Z and M. As an example, this is how you might insert a multipoint with 4 parts for each point:

INSERT INTO geotable ( geom ) 
  VALUES ( GeomFromEWKT('SRID=4326;MULTIPOINT(-126.4 45.32 15 17.2,-125.1 44.2 15.1 14.9)') )

Your Answer

By clicking "Post Your Answer", you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.