From: | "Ramya Chandrasekar" <csekar(dot)ramya(at)gmail(dot)com> |
---|---|
To: | "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Regd: TODO Item : Have EXPLAIN ANALYZE issue NOTICE messages ... |
Date: | 2008-07-27 03:45:30 |
Message-ID: | e3fd59f00807262045g4a3cf7a6q995ab26661331643@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Regarding: Optimizer To Do Item: Have EXPLAIN ANALYZE issue NOTICE messages
when the estimated and actual row counts differ by a specified percentage.
Hi,
After going through the thread related to the above mentioned to do Item, my
understanding is that:
Issuing notices when a problematic node occurs is useful in the case of long
running queries. It can avoid having the user wait a couple of hours to
solve the problem.
A possible solution from the archives was summarized as below:
Bruce Momjian Wrote :
I was thinking we could issue NOTICE when the estimates differed from
the actual by a specified percentage, and that NOTICE could be issued
while the query is still processing, assuming the stage completes before
the query does. This seems much easier than doing protocol changes.
Looking forward to your inputs on this.
Thank you,
Ramya
On Thu, Jun 26, 2008 at 10:22 AM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> "Ramya Chandrasekar" <csekar(dot)ramya(at)gmail(dot)com> writes:
> > Regd: TODO Item: Optimizer:: Have EXPLAIN ANALYZE issue NOTICE messages
> when
> > the estimated and actual row counts differ by a specified percentage.
>
> I think that TODO item hasn't been thought through very carefully.
> NOTICEs that are dissociated from the actual explain output seem pretty
> useless, or at least painful to use. You might want to go back in the
> archives and read the whole thread that led up to the TODO item to see
> what ideas were kicked around.
>
> regards, tom lane
>
From | Date | Subject | |
---|---|---|---|
Next Message | Tatsuo Ishii | 2008-07-27 08:47:11 | Re: WITH RECUSIVE patches 0723 |
Previous Message | Andrew Dunstan | 2008-07-27 03:03:51 | Re: Search for version-numbered tclsh? |