Re: performance and number of selected columns

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Dirk Lutzebaeck <lutzeb(at)aeccom(dot)com>
Cc: pgsql-general(at)hub(dot)org
Subject: Re: performance and number of selected columns
Date: 2000-05-22 15:58:06
Message-ID: 18117.959011086@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Dirk Lutzebaeck <lutzeb(at)aeccom(dot)com> writes:
> say I have three tables a,b,c with lots of columns (say 10 for each
> table). Using psql command line, why does
> SELECT a.*,b.*,c.* FROM a,b,c
> takes much more longer (in my specifc case 3 times) than only
> selecting one column like
> SELECT a.oid, b.oid, c.oid FROM a,b,c.

I'd guess you are simply looking at the increased time to transfer
the additional data to the frontend, format it for display, etc.
With a pure cartesian-product join like that (no WHERE), the backend
is basically going to be spitting out tuples as fast as it can read
them from disk...

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Roderick A. Anderson 2000-05-22 16:40:15 Re: RPM troubleshoot
Previous Message Ron Peterson 2000-05-22 15:57:03 Re: Re: [HACKERS] Postgresql OO Patch