Re: BUG #6275: Horrible performance regression - Mailing list pgsql-bugs

From Guillaume Smet
Subject Re: BUG #6275: Horrible performance regression
Date
Msg-id CALt0+o-L+2ajCSbFQQ-68=aYj-hNkEGw0baE4p_nBOSFxyKoyA@mail.gmail.com
Whole thread Raw
In response to Re: BUG #6275: Horrible performance regression  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-bugs
On Mon, Oct 31, 2011 at 1:50 PM, Robert Haas <robertmhaas@gmail.com> wrote:
> If you can't do that for some reason, there's always auto_explain, but
> that has some overhead and is a bit more work to set up.

If it's a massive load of data with a lot of queries, I think his best
bet is to log the queries with log_min_duration_statement = <something
sensible> (I'd try to start with 30ms) then analyze the log results
with pgFouine or any other log analyzer for both versions.

--
Guillaume

pgsql-bugs by date:

Previous
From: Tom Lane
Date:
Subject: Re: BUG #6278: Index scans on '>' condition on field with many NULLS
Next
From: Robert Haas
Date:
Subject: Re: BUG #6278: Index scans on '>' condition on field with many NULLS