[postgis-users] Improving Performance

[Wood Brent]

--- Ralph Mason <ralph.mason at telogis.com> wrote:
> David Blasby wrote:
> 
> > Ralph Mason wrote:
> >
> >> I think it wouldn't be to difficult to benchmark the bounding box 
> >> being floats / doubles / gone altogether.  The optimum setting is 
> >> likely to be based on the actual machine it's running on.  I wouldn't 
> >> be surprised if no bounding box was actually the fastest on a modern 
> >> fast processor with a good sized cache.

I am curently using 2 main types of point-in-poly query, one is against a table
of cells (boxes). The bounding box test is irrelevant, and if it tests the box
in the index, then the geometry, the test is performed twice for each hit as
the box IS the feature.

The second type of query is against coastlines to identify points on/off land.
The polygons in this case may have millions of vertices & the bounding box test
is enormously faster.

A standard approach to geometries (& indexing), instead of distinct geometric
types (& specialist indexes) may not be the most efficient approach to handling
the data. I'm not sure how PostGIS treats different types of geometries, or how
I can optimise it for such situations.


Brent Wood