Re: Uh, I change my mind about commit_delay + commit_siblings (sort of)

On 31 May 2012 15:00, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Simon Riggs <simon(at)2ndQuadrant(dot)com> writes:
>> On 31 May 2012 13:16, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>>> Frankly, I think this whole thing should be pushed to 9.3.
>
>> What matters is that we have a patch that provides a massive
>> performance gain in write performance in just a few lines of code, and
>> that should be committed to 9.2.
>
> I agree with Robert on this.  This patch hasn't had *nearly* enough
> testing to justify cramming it into 9.2 at this point.  AFAIK the
> claim of "massive performance gain" is based on a single test case run
> by a single person, which doesn't even give me any confidence that it
> doesn't break anything, much less that it's a win across the board.

I agree with you. You would be mistaken if you thought that I think
Peter's laptop was sufficient proof for anyone to commit something and
I've already said exactly that to him.

My description of "massive performance gain" is appropriate based on
the measurements so far.

> If we want to finish the beta cycle in a reasonable time period and get
> back to actual development, we have to refrain from adding more
> possibly-destabilizing development work to 9.2.  And that is what
> this is.

In what way is it possibly destabilising? I see nothing in the patch
to merit that claim, so presumably you haven't read the patch yet?

--
 Simon Riggs                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services