On 22/11/2009 11:52, Thom Brown wrote:
> Since this is a performance issue, this should probably have been sent
> to the pgsql-performance mailing list. But in any case, the vast
> majority of performance issues require an EXPLAIN output, or preferably
> with ANALYZE also as there is nothing to help diagnose what the query
> planner it attempting to do.
You may think it's a performance issue, but it stems from the decision
to remove from Postgres an essential facility, that of generating and
using R-trees.
I'm currently trying to generate test cases, together with EXPLAIN output.
Regards
--
Clive Page