: Ozer, Pam
Cc: Josh Berkus; pgsql-performance@postgresql.org
Subject: Re: [PERFORM] Slow Query- Bad Row Estimate
"Ozer, Pam" writes:
> Yes. The default statistics target was at 1000. So that would be
what the column was using correct?
But you evidently didn't have stats.
"Ozer, Pam" writes:
> Yes. The default statistics target was at 1000. So that would be what the
> column was using correct?
But you evidently didn't have stats. Perhaps you have autovacuum turned
off? What PG version is this anyway?
regards, tom lane
--
Sent via pg
Query- Bad Row Estimate
On 10/29/10 2:47 PM, Ozer, Pam wrote:
> I had just analyzed the dealergroupgeochache table. Wow. Thank you. That did
> the trick. Can you give me an explanation of the default_stats work? I don't
> think I completely understand what it means when you set it
On 10/29/10 2:47 PM, Ozer, Pam wrote:
> I had just analyzed the dealergroupgeochache table. Wow. Thank you. That did
> the trick. Can you give me an explanation of the default_stats work? I don't
> think I completely understand what it means when you set it to 500 instead of
> 1000?
You're al
"Ozer, Pam" writes:
> I am not sure what you mean by reformulate the data representation. Do
> you mean do I have to join on all three columns?
No, I was wondering if you could change things so that you join on just
one column, instead of two that each tell part of the truth.
BTW, did you chec
pgsql-performance-ow...@postgresql.org
[mailto:pgsql-performance-ow...@postgresql.org] On Behalf Of Josh Berkus
Sent: Friday, October 29, 2010 2:10 PM
To: pgsql-performance@postgresql.org
Subject: Re: [PERFORM] Slow Query- Bad Row Estimate
On 10/29/10 1:54 PM, Ozer, Pam wrote:
> "
: [PERFORM] Slow Query- Bad Row Estimate
"Ozer, Pam" writes:
> Unfortunately I have not received a response on this question. Is
more
> information needed? Does anyone have any ideas why the estimates may
be
> bad? Or what I might be able to do to speed this up?
The most lik
"Ozer, Pam" wrote:
> Is more information needed?
Table layouts of the tables involved (including indexes) would be
interesting. A description of the machine would be useful,
including OS, CPUs, RAM, and disk system.
I know you said you might have trouble changing the config, but some
of the
"Ozer, Pam" writes:
> Unfortunately I have not received a response on this question. Is more
> information needed? Does anyone have any ideas why the estimates may be
> bad? Or what I might be able to do to speed this up?
The most likely explanation for the bad rowcount estimates is that there
On 10/29/10 1:54 PM, Ozer, Pam wrote:
> "-> Index Scan using dealergroupgeocache_i01 on
> dealergroupgeocache (cost=0.00..5719.56 rows=9055 width=10) (actual
> time=0.015..87.689 rows=163491 loops=1)"
This appears to be your problem here.
a) when was dealergroupgeocache last
10 matches
Mail list logo
