Well, heh I'm no SQL expert. I kinda piece things together the best I can
from what I can read and this was really the only way I could make the
UPDATE work correctly. But the plan looks complicated with a lot of hash
conditions, hash joins, and scans. I'm worried it wont perform with a very
large
> -Original Message-
> From: pgsql-performance-ow...@postgresql.org [mailto:pgsql-
> performance-ow...@postgresql.org] On Behalf Of Kevin Grittner
> Sent: Friday, August 09, 2013 11:44 AM
> To: Robert DiFalco; pgsql-performance@postgresql.org
> Subject: Re: [PERFORM] Efficient Correlated Up
I sometimes experience that updating smaller sets is more efficient than
doing all at once in one transaction (talking about 1+)
Always make sure the update references can make use of indices
On Fri, Aug 9, 2013 at 5:44 PM, Kevin Grittner wrote:
> Robert DiFalco wrote:
>
> > In my system
Robert DiFalco wrote:
> In my system a user can have external contacts. When I am
> bringing in external contacts I want to correlate any other
> existing users in the system with those external contacts. A
> users external contacts may or may not be users in my system. I
> have a user_id field i
