On Wed, 2008-07-23 at 10:57 +0800, Xiao Meng wrote:
> Well, I'll do it after I finish my second patch.
> Hash index should be more efficient than btree when N is big enough.
> It seems meaningful to find how big N is in an experiment way.
Agreed.
We should also examine the basic thinking of the index.
My understanding is that it dynamically resizes hash as the index grows.
If we already believe the only benefit would come when the index is
large, having special handling for small tables seems like a waste of
time because we will never use it in those contexts.
So starting the hash at a fairly large size makes more sense than it
might otherwise seem to.
-- Simon Riggs www.2ndQuadrant.comPostgreSQL Training, Services and Support