Re: Query length restriction in v3 protocol?

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Mark,

Mark Lewis schrieb am 11.05.2006 00:33:
> Patric,
>
>> Hi Mark,
>>
> The outer join improvements are already in CVS HEAD:
>
> http://archives.postgresql.org/pgsql-performance/2006-02/msg00342.php
>
> If you test with CVS HEAD could you please post your results back here?
> I know I'd be really interested in the results.
>
from the mail: "
CVS HEAD can re-order left joins in common cases, but no existing
release will touch the ordering of outer joins at all.
"

That does not sound like there's anything done with the outer joins...?
I want to be sure there's a realistic chance of enhancing things
really.. maybe you have a link onto the CVS commit info?

So far, I even disabled the nestedloop on all 8.1 servers and increased
effective_cache_size to push the optimizer to do index lookups more
often in that case. So far, it helps... but all in all this wasn't
necessary in 8.0...
But nevertheless, the union mapping seems *far* better performance-wise
atm. Pity. No constraints. No unique id index. ;-(

> Ah, that's too bad.  The one place where we've needed to care about
> Hibernate's join order was at a place where our inheritance tree is
> flat, which explains why this trick worked for us.

Yep. That's why it works for you... We have the common superclass
pattern in our ORM, which I like most, and our inheritance trees are
partly very deep (8-10 levels is quite common) and make heavy use of
polymorphic relations and queries. Which is wonderful and elegant,
object-wise seen. Which is a great pita on the SQL side... :-\

- --
Mit freundlichen Gruessen / Regards
Patric Bechtel, IPCON Informationssysteme OHG
Kontakt: http://www.ipcon.de/kontakt.php
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (MingW32)
Comment: GnuPT 2.5.2

iD8DBQFEYqDwfGgGu8y7ypARAhLfAKCvG8bVSyy12R8jT9EInLVbgeMDpACeO3Yl
VjgWeaR5WMyI0gJ8tryfgWY=
=pHBZ
-----END PGP SIGNATURE-----