Any way to optimize GROUP BY queries? - Mailing list pgsql-performance

From Cristian Prieto
Subject Any way to optimize GROUP BY queries?
Date
Msg-id 00d701c604c1$e6945d80$6500a8c0@gt.ClickDiario.local
Whole thread Raw
Responses Re: Any way to optimize GROUP BY queries?  (Greg Stark <gsstark@mit.edu>)
List pgsql-performance

I have the following table:

 

CREATE TABLE mytmp (

            Adv integer,

            Pub integer,

            Web integer,

            Tiempo timestamp,

            Num integer,

            Country varchar(2)

);

 

CREATE INDEX idx_mytmp ON mytmp(adv, pub, web);

 

And with 16M rows this query:

 

SELECT adv, pub, web, country, date_trunc(‘hour’, tiempo), sum(num)

FROM mytmp GROUP BY adv, pub, web, country, date_trunc(‘hour’, tiempo)

 

I’ve tried to create index in different columns but it seems that the group by clause doesn’t use the index in any way.

 

Is around there any stuff to accelerate the group by kind of clauses?

 

Thanks a lot…

 

 

 

Attachment

pgsql-performance by date:

Previous
From: "Luke Lonergan"
Date:
Subject: Re: PostgreSQL and Ultrasparc T1
Next
From: James Klo
Date:
Subject: Re: make bulk deletes faster?