Re: Review remove {join,from}_collapse_limit, add enable_join_ordering

On Thu, Jul 16, 2009 at 4:32 PM, Tom Lane<tgl@sss.pgh.pa.us> wrote:
> samples  %        image name               symbol name
> 886498   53.8090  postgres                 have_relevant_eclass_joinclause
> 460596   27.9574  postgres                 bms_overlap
>
> So maybe a redesign of the equivalence-class joinclause mechanism is in
> order.  Still, this is unlikely to fix the fundamental issue that the
> time for large join problems grows nonlinearly.

Perhaps it's GEQO's fault that it's using these functions
inappropriately, calling them often to calculate these answers
whenever it needs them instead of looking once for join clauses and
then optimizing based on the results. But I've never actually looked
at geqo, mabe that's inherent in the design?


--
greg
http://mit.edu/~gsstark/resume.pdf