it was not willing :-)
it just stayed so after various probes with a query plan.
Anyway, on 8.4 the default target is 100, an just by move it to 5 I
reached on 16cores 10.500 TPS instead of 8.000 initially. And I think
you have a good reason to keep it equal to 100 by default, isn't it?
And what about scalability on 32cores?..
On 5/11/09, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Dimitri <dimitrik(dot)fr(at)gmail(dot)com> writes:
>> Anyone may explain me why analyze target may have so huge negative
>> secondary effect?..
> If these are simple queries, maybe what you're looking at is the
> increase in planning time caused by having to process 10x as much
> statistical data. Cranking statistics_target to the max just because
> you can is not necessarily a good strategy.
> regards, tom lane
In response to
pgsql-performance by date
|Next:||From: Dimitri||Date: 2009-05-11 16:31:58|
|Subject: Re: What is the most optimal config parameters to keep stable write TPS ?..|
|Previous:||From: Kevin Grittner||Date: 2009-05-11 16:19:39|
|Subject: Re: What is the most optimal config parameters tokeep stable write TPS ?..|