Skip site navigation (1) Skip section navigation (2)

Re: [PERFORM] Proposal: relaxing link between explicit JOINs and execution order

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: josh(at)agliodbs(dot)com
Cc: pgsql-hackers(at)postgreSQL(dot)org, pgsql-performance(at)postgreSQL(dot)org
Subject: Re: [PERFORM] Proposal: relaxing link between explicit JOINs and execution order
Date: 2003-01-23 00:21:24
Message-ID: 25650.1043281284@sss.pgh.pa.us (view raw or flat)
Thread:
Lists: pgsql-hackerspgsql-performance
Josh Berkus <josh(at)agliodbs(dot)com> writes:
> I am very strongly in favor of this idea.   I would personally prefer it if 
> the Join collapsing parmeter could be set at query time through a SET 
> statement, but will of course defer to the difficulty level in doing so.

I guess I failed to make it clear that that's what I meant.  GUC
variables are those things that you can set via SET, or in the
postgresql.conf file, etc.  These values would be just as manipulable
as, say, ENABLE_SEQSCAN.

> How about:
> EXPLICIT_JOIN_MINIMUM
> and
> FROM_COLLAPSE_LIMIT

> Just to make the two params not sound so identical?

Hmm.  The two parameters would have closely related functions, so I'd
sort of think that the names *should* be pretty similar.

			regards, tom lane

In response to

pgsql-performance by date

Next:From: Dann CorbitDate: 2003-01-23 00:22:51
Subject: Re: Terrible performance on wide selects
Previous:From: Dann CorbitDate: 2003-01-23 00:21:18
Subject: Re: Terrible performance on wide selects

pgsql-hackers by date

Next:From: Dann CorbitDate: 2003-01-23 00:22:51
Subject: Re: Terrible performance on wide selects
Previous:From: Dann CorbitDate: 2003-01-23 00:21:18
Subject: Re: Terrible performance on wide selects

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group