| From: | Craig Ringer <craig(at)postnewspapers(dot)com(dot)au> |
|---|---|
| To: | Carlo Stonebanks <stonec(dot)register(at)sympatico(dot)ca> |
| Cc: | pgsql-performance(at)postgresql(dot)org |
| Subject: | Re: Function scan/Index scan to nested loop |
| Date: | 2010-05-11 07:14:08 |
| Message-ID: | 4BE903C0.4020203@postnewspapers.com.au |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On 11/05/10 13:32, Carlo Stonebanks wrote:
> Hello all,
>
> A query ran twice in succession performs VERY poorly the first time as
> it iterates through the nested loop. The second time, it rips. Please
> see SQL, SLOW PLAN and FAST PLAN below.
I haven't looked at the details, but the comment you made about it being
fast on the live server which hits this query frequently tends to
suggest that this is a caching issue.
Most likely, the first time Pg has to read the data from disk. The
second time, it's in memory-based disk cache or even in Pg's
shared_buffers, so it can be accessed vastly quicker.
--
Craig Ringer
Tech-related writing: http://soapyfrogs.blogspot.com/
| From | Date | Subject | |
|---|---|---|---|
| Next Message | venu madhav | 2010-05-11 08:47:57 | Performance issues when the number of records are around 10 Million |
| Previous Message | Scott Marlowe | 2010-05-11 07:07:26 | Re: Function scan/Index scan to nested loop |