| From: | Eduard Català <eduard(dot)catala(at)gmail(dot)com> |
|---|---|
| To: | pgsql-performance(at)lists(dot)postgresql(dot)org |
| Subject: | Table order at FROM clause affects performance? |
| Date: | 2018-04-12 14:35:09 |
| Message-ID: | CAL54xNYE38bzJGKWbhAe3g_36yc8HNLhxwPJRqs85qdKGAxEmQ@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
I have a strange performance issue, i think that is not possible:
Given this statement:
SELECT *several_fields* FROM A, B, C WHERE *conditions*
A, B are tables with several LEFT JOINS but they act as one subquery.
If I execute the select above:
SELECT *several_fields* FROM A, B, C WHERE *conditions*
*Time: 30 secs*
*Cost: 1M*
If I execute the same select (same parameters) but swapping A and B in the
from clause:
SELECT *several_fields* FROM B, A, C WHERE *conditions*
*Time: 19ms*
*Cost: 10k*
The plan changes dramatically: I can't see why the order of FROM clause
impacts directly on the query cost and plan. If this is possible, where i
can read about it? I need to know how the order of FROM clause modifies the
query plan.
Thanks in advance. This is my first post.
Eduard Català
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2018-04-12 15:30:28 | Re: Table order at FROM clause affects performance? |
| Previous Message | Mark Kirkwood | 2018-04-12 05:11:08 | Re: Latest advice on SSD? |