Hi All,
My question for this week:
How far is the above subject from beeing implemented?
(I've run through the archives, and found some quite old mails in
this topic. They mentioned automatically generated temp tables as possibe
solutions... Is this the plan even today?)
Attila
Ps.:
Two weeks ago I had a question about multiplee joins. According to Tom's
advice I translated the actual snapshot from cvs and after submitting the
query with parenthesis on the right place I experienced a speedup factor
of 200 (two hundred!!!) comapred to the GEQO. Great.
Ps^2.:
Also tested tuple toasting up to 4Mbytes of single tuples...
I like it :))