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

On 31 May 2012 16:26, Peter Geoghegan <peter(at)2ndquadrant(dot)com> wrote:
> On 31 May 2012 16:23, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Simon Riggs <simon(at)2ndQuadrant(dot)com> writes:
>>> In what way is it possibly destabilising?
>>
>> I'm prepared to believe that it only affects performance, but it could
>> be destabilizing to that.  It needs proper review and testing, and the
>> next CF is the right environment for that to happen.
>
> It couldn't possibly be as destabilising to performance as
> commit_delay was in 9.1.

Furthermore, it couldn't possibly affect performance in any way unless
commit_delay is set. I've just moved the delay site so that its only
executed by the group commit leader. The leader would execute the code
anyway, but now the followers don't.

--
Peter Geoghegan       http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training and Services