Re: Large tables, ORDER BY and sequence/index scans - Mailing list pgsql-general

From Milan Zamazal
Subject Re: Large tables, ORDER BY and sequence/index scans
Date
Msg-id 87k4vw4m8q.fsf@blackbird.nest.zamazal.org
Whole thread Raw
In response to Re: Large tables, ORDER BY and sequence/index scans  ("Albe Laurenz" <laurenz.albe@wien.gv.at>)
List pgsql-general
>>>>> "AL" == Albe Laurenz <laurenz.albe@wien.gv.at> writes:

    AL> Did you try to reduce the cursor_tuple_fraction parameter?

No, good idea, thanks.  It helps.

The question is whether it's a good idea to reduce cursor_tuple_fraction
universally, without knowing the table size before (and I'm not going to
use SELECT COUNT(*) for well known reasons).  But I can probably
experiment a bit and will see, it looks promising.


pgsql-general by date:

Previous
From: Milan Zamazal
Date:
Subject: Re: Large tables, ORDER BY and sequence/index scans
Next
From: Milan Zamazal
Date:
Subject: Re: Large tables, ORDER BY and sequence/index scans