Re: Hash join order

"PdV" <piet.de.visser_at_cmg.com> wrote in message news:6de2ae57.0303310740.61a76ff8_at_posting.google.com...
> "Telemachus" <telemachus_at_ulysseswillreturn.net> wrote in message
> > Note this is probably not a definitive list; new patches to oracle may
> > introduce more new and undocumented hints. Also, this may not be all the
> > hints there either. YMMV.
> > Also the list of documented hints may also shrink as the optimizer gets
> > better and better at figuring stuff out.

<aside to telemachus>
don't bet on it
</aside>

> 2. to properly use the most outrageous options such /* +_MAJOR_TOM */
> /* +_EMMINEM */, or /* +EVEN_FASTER */
> I would also need some documented insight in the workings of
> any particular undocumented hint.

Well yes and no. For example INDEX_SS is listed as undocumented (which it is). Yet the index skip scan operation *is* documented (at http://download-west.oracle.com/docs/cd/B10501_01/server.920/a96533/optimops .htm#51553 ). If one can work out what the hint is achieving one may well find the requisite considerations documented. Of course this is close to saying 'well if you know what you are looking at you can see what you are looking at' but I don't think it is quite the same.

--
Niall Litchfield
Oracle DBA
Audit Commission UK
*****************************************
Please include version and platform
and SQL where applicable
It makes life easier and increases the
likelihood of a good answer
******************************************