[postgis-users] Why the creation of an index GIST is much faster thanthe creation of an index RTREE

[david.garnier at etudier-online.com]

> 
> It seems to me that not only do GIST and R-tree differ in approach
> (lossiness, the stats they gather)
> 
>  but that GIST is simply more fully implemented in postgres.
> 
> Hope this is of use,
> 
> Rob

I had at look the definition of R-tree and GiST, and basically a GiST is a 
generic tree structure that can be used to implement, guess what, R-trees and B-
trees. So when you use a GiST index in Postgis, you basically use an R-tree. I 
think that the difference in speed comes from the fact that the new 
implementation of the R-tree using the GiST is better than the older, native 
one.

Best regards,
David Garnier