Re: parameter hints to the optimizer

From: Hans-Jürgen Schönig <postgres(at)cybertec(dot)at>
To: Merlin Moncure <merlin(dot)moncure(at)rcsonline(dot)com>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: parameter hints to the optimizer
Date: 2004-07-21 19:30:22
Message-ID: 40FEC44E.6030504@cybertec.at
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Merlin Moncure wrote:
>>Merlin,
>>
>>This will most likely never be accepted by the core team because it is
>>better to spend more time on fixing the planner than to invent some
>>non-standard.
>>As far as I know some databases support a syntax where hints can be
>>hidden in comments or something like that.
>>Meanwhile I think that hints are more of a burdon than a help.
>>
>> Regards,
>>
>> Hans
>
>
> I thought as much. Still, the parameter problem is a huge headache. Maybe if it was possible to use the statistics to gather a 'pseudovariable' to feed to the parameter based on some algorithm, the planner could give better results without exposing the planner inner workings to the user.
>
> Thanks for the feedback.
> Merlin

Using statistics is exactly what the planner does ...
So why should data coming from the planner being given back to the
planner? Doesn't make sense.

For more information I highly recommend Tom's talk at Oreilly's some
time ago. I think it is called "recent improvements in 7.4" ...

regards,

Hans

--
Cybertec Geschwinde u Schoenig
Schoengrabern 134, A-2020 Hollabrunn, Austria
Tel: +43/720/10 1234567 or +43/664/816 40 77
www.cybertec.at, www.postgresql.at, kernel.cybertec.at

Browse pgsql-hackers by date

  From Date Subject
Next Message Simon Riggs 2004-07-21 20:00:48 PreallocXlogFiles
Previous Message Dann Corbit 2004-07-21 18:36:47 miniscule compiler barf in pg_ctl.c