| From: | Robert Haas <robertmhaas(at)gmail(dot)com> |
|---|---|
| To: | Mark Thornton <mthornton(at)optrak(dot)co(dot)uk> |
| Cc: | pgsql-performance(at)postgresql(dot)org |
| Subject: | Re: Slow join on partitioned table |
| Date: | 2011-03-04 18:06:25 |
| Message-ID: | AANLkTi=ZJziFyG4v23grw-gzapFB=uR=yEPzV1DpqhyD@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On Fri, Mar 4, 2011 at 12:00 PM, Mark Thornton <mthornton(at)optrak(dot)co(dot)uk> wrote:
> On 04/03/2011 16:07, Robert Haas wrote:
>>
>> That seems quite surprising. There are only 14 rows in the table but
>> PG thinks 2140? Do you have autovacuum turned on? Has this table
>> been analyzed recently?
>>
> I think autovacuum is enabled, but as a temporary table LinkIds has only
> existed for a very short time (at least in my current tests).
Autovacuum doesn't work on temporary tables, so time of existence
doesn't matter in that case. The best approach is to load the data,
then analyze, then query it.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Dan Harris | 2011-03-04 18:39:56 | Re: Linux I/O schedulers - CFQ & random seeks |
| Previous Message | Wayne Conrad | 2011-03-04 18:03:36 | Re: Linux I/O schedulers - CFQ & random seeks |