Hi Lenz,
> > Part of the problem (at least where myisam is involved) is the way the
> > optimizer calculates the cost.
> >
> > 2 older patches to gain about 20% in the myisam key lookup are
> > http://bugs.mysql.com/bug.php?id=26232
> > http://bugs.mysql.com/bug.php?id=26453
> >
> > This isn't enough of course. and I am looking fwrd to see how this may be
> > solved. (Also the above affects myisam only)
>
> Well, gaining 20% is not too shabby!
>
> I notice that there has not been much activity on getting these patches
> incorporated since Martin posted them in February this year. Sergey, as you
> are the assignee for both of these: what's the status? Any plans on incorporating
> Martin's suggestions?
Unfortunately so far I've only suceeded at keeping them in my bugs
queue. At every point in time there was some more burning issue. From
quick look at the patches I don't have any objections, but there is a
need for further study. I hope to get to them at some point but I
don't have an ETA. Since the patches touch only code within MyISAM,
perhaps we could try to see if Engines team has time to handle them.
BR
Sergey
--
Sergey Petrunia, Software Developer
MySQL AB, www.mysql.com
Office: N/A
Blog: http://s.petrunia.net/blog
