Re: order by slowing down a query by 80 times - Mailing list pgsql-performance

From Rajesh Kumar Mallah
Subject Re: order by slowing down a query by 80 times
Date
Msg-id AANLkTimScoXZXF9Dq797f611P6qi4MRY3SVPshDPLSpK@mail.gmail.com
Whole thread Raw
In response to Re: order by slowing down a query by 80 times  (Yeb Havinga <yebhavinga@gmail.com>)
List pgsql-performance


On Mon, Jun 28, 2010 at 5:09 PM, Yeb Havinga <yebhavinga@gmail.com> wrote:
Rajesh Kumar Mallah wrote:
Dear List,

just by removing the order by co_name reduces the query time dramatically
from  ~ 9 sec  to 63 ms. Can anyone please help.
The 63 ms query result is probably useless since it returns a limit of 25 rows from an unordered result. It is not surprising that this is fast.

The pain is here:

Index Scan using profile_master_co_name on profile_master b  (cost=0.00..1125295.59 rows=6968 width=25) (actual time=0.097..9193.154 rows=2212 loops=1)
                               Filter: ((co_name IS NOT NULL) AND ((co_name_vec)::tsvector @@ to_tsquery('manufacturer'::text)))


It looks like seq_scans are disabled, since the index scan has only a filter expression but not an index cond.


seq_scans is NOT explicitly disabled. The two queries just differed in the order by clause.

regds
Rajesh Kumar Mallah.
 

regards,
Yeb Havinga



Regds
Rajesh Kumar Mallah.

pgsql-performance by date:

Previous
From: Andres Freund
Date:
Subject: Re: order by slowing down a query by 80 times
Next
From: "Kevin Grittner"
Date:
Subject: Re: order by slowing down a query by 80 times