Bryan Pendleton wrote:
Independently, I tried fiddling with the cost estimates myself, because as
I stepped through HeapCostController and BTreeCostController, my reaction
was that the costs were possibly 2 orders of magnitude too high. In my
experiments, I didn't see much benefit to the optimizer, but I wonder if
that's because I *also* need to have some of these other fixes (such as the
permute state fix from jumpReset.patch and the JUMPING-give-up change that
you've been studying) in order for the cost estimate changes to take
effect?
I was actually going a very different direction with this. I don't know
anything about the CostControllers so I tend to avoid them, perhaps to my
detriment ;)
I've attached a partial patch with my (potentially completely wrong) approach to
DERBY-1905. Feel free to look at that and comment/fix it up as needed. I don't
have time at the moment to pursue those changes and will not be able to until at
least next year...
Army
