[postgis-users] Efficiency of <->, # and &&/st_distance

[Jan Hartmann]

Hi,

I am working on an interactive application to add and remove points from 
a map. When removing, the user clicks on the map and should get back the 
nearest point in the database. Normally, I would do this with something 
like: "select ... where <point> && the_geom and 
st_distance(<point>,the_geom) < 10)". I see there are two new operators 
now: # and <->. Is there any difference in efficiency in this case?

Also: all three methods require a GIST index. However, this is a very 
interactive application that continuously changes the number of points. 
The number of points can potentially grow quite large (hundreds of 
thousands) Are there any rules to organise the database so as to 
optimize GIST access and update (e.g. maximum number of records, 
partitioning into multiple tables)?

Jan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-users/attachments/20120206/a2de7c05/attachment.html>