Re: Shouldn't we have a way to avoid "risky" plans? - Mailing list pgsql-performance

From Tom Lane
Subject Re: Shouldn't we have a way to avoid "risky" plans?
Date
Msg-id 4286.1300914004@sss.pgh.pa.us
Whole thread Raw
In response to Re: Shouldn't we have a way to avoid "risky" plans?  (Claudio Freire <klaussfreire@gmail.com>)
Responses Re: Shouldn't we have a way to avoid "risky" plans?
Re: Shouldn't we have a way to avoid "risky" plans?
Re: Shouldn't we have a way to avoid "risky" plans?
List pgsql-performance
Claudio Freire <klaussfreire@gmail.com> writes:
> On Wed, Mar 23, 2011 at 5:29 PM, Josh Berkus <josh@agliodbs.com> wrote:
>> On 3/23/11 10:35 AM, Claudio Freire wrote:
>>> �* �consider plan bailout: execute a tempting plan, if it takes too
>>> long or its effective cost raises well above the expected cost, bail
>>> to a safer plan

>> That would actually solve this particular case. �It would still require
>> us to have some definition of "safer" though.

> In my head, safer = better worst-case performance.

If the planner starts operating on the basis of worst case rather than
expected-case performance, the complaints will be far more numerous than
they are today.

            regards, tom lane

pgsql-performance by date:

Previous
From: Justin Pitts
Date:
Subject: Re: Shouldn't we have a way to avoid "risky" plans?
Next
From: Claudio Freire
Date:
Subject: Re: Shouldn't we have a way to avoid "risky" plans?