Date: Wed, 5 Dec 2007 13:36:58 -0300
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: "Joshua D. Drake" <jd@commandprompt.com>
Cc: Greg Sabino Mullane <greg@turnstep.com>, pgsql-admin@postgresql.org
Subject: Re: Enhancement request
Message-ID: <20071205163658.GE16991@alvh.no-ip.org>
References: <09fbbcab762fc5279e45349ce1cbad93@biglumber.com>
	<4756CB9B.3050301@commandprompt.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <4756CB9B.3050301@commandprompt.com>
User-Agent: Mutt/1.5.16 (2007-06-11)

Joshua D. Drake wrote:
> Greg Sabino Mullane wrote:

>> I am curious however, as to what the future vision of autovacuum is.
>> Will it be so efficient that it won't impact all but the busiest tables? 
>> Will it be able to figure out the best time to vacuum on its own?
>
> I believe maintenance windows are on the horizon. Of course it seems that 
> instead of focusing on that, we should just put in a generic job manager.

When I proposed my idea of maintenance window, everybody seemed happy,
so that's what got implemented.

If you want to propose a generic job manager, be my guest.


>> Will it stop itself mid-run if the need arises?
>
> Man I would love this one... coupled with a just enough vacuum. E.g; it 
> calculates a percentage it needs to vacuum, and only vacuums that much. 
> Leaving dead tuples behind.

Stop dreaming and start assigning resources for it to get done ;-)  The
NTT guys have done so and they are ahead of us on that camp.

-- 
Alvaro Herrera                          Developer, http://www.PostgreSQL.org/
"In Europe they call me Niklaus Wirth; in the US they call me Nickel's worth.
 That's because in Europe they call me by name, and in the US by value!"