From: | Csaba Nagy <nagy(at)ecircle-ag(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Merlin Moncure <merlin(dot)moncure(at)rcsonline(dot)com>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: parameterized limit statements |
Date: | 2005-11-08 09:48:28 |
Message-ID: | 1131443308.4892.61.camel@coppola.muc.ecircle.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Mon, 2005-11-07 at 18:43, Tom Lane wrote:
[snip]
> If it doesn't have a value for the parameter, it'll assume 10% of table
> rows, which is what it's done for a long time if the LIMIT isn't
> reducible to a constant.
Is 10% a reasonable guess here ?
Here we use limit in combination with prepared statements to get
something like less than 1% of the table. There are no exceptions to
that in our code... even if the limit amount is a parameter.
Furthermore, the limit amount is always a small number, usually ~ 100,
but never more than 1000. So in my case, we could live with a suboptimal
plan when the percentage would be more than 10%, cause then the table
would be small enough not to matter that much. In turn it has a huge
impact to wrongly guess 10% for a huge table...
I think the best would be to guess 5% but maximum say 5000. That could
work well with both small and huge tables. Maybe those values could be
made configurable... just ideas, not like I could implement this...
[snip]
Cheers,
Csaba.
From | Date | Subject | |
---|---|---|---|
Next Message | Teodor Sigaev | 2005-11-08 12:13:32 | Re: SIGSEGV taken on 8.1 during dump/reload |
Previous Message | Sreejesh O S | 2005-11-08 09:22:10 | Is there any other way to compile pgsql without gmake |