Re: Re: Increasing work_mem and shared_buffers on Postgres 9.2 significantly slows down queries

Поиск
Список
Период
Сортировка
От Gunnar \"Nick\" Bluth
Тема Re: Re: Increasing work_mem and shared_buffers on Postgres 9.2 significantly slows down queries
Дата
Msg-id 50962BCB.7020709@pro-open.de
обсуждение исходный текст
Ответ на Increasing work_mem and shared_buffers on Postgres 9.2 significantly slows down queries  (Petr Praus <petr@praus.net>)
Ответы Re: Re: Increasing work_mem and shared_buffers on Postgres 9.2 significantly slows down queries
Список pgsql-performance
Am 03.11.2012 18:19, schrieb Petr Praus:
On 3 November 2012 12:09, Gunnar "Nick" Bluth <gunnar.bluth@pro-open.de> wrote:
Am 03.11.2012 16:20, schrieb Petr Praus:

Your CPUs are indeed pretty oldschool. FSB based, IIRC, not NUMA. A process migration would be even more expensive there.


Ok, I've actually looked these up now... at the time these were current, I was in the lucky situation to only deal with Opterons. And actually, with these CPUs it is pretty possible that Scott Marlowe's hint (check vm.zone_reclaim_mode) was pointing in the right direction. Did you check that?

I did check that, it's zero. I responded to his message, but my messages to the mailing list are getting delayed by ~24 hours because somebody has to always bless them.
 


Yes, same behaviour. I let the shared_buffers be the default (which is 8MB). With work_mem 1MB the query runs fast, with 96MB it runs slow (same times as before). It really seems that the culprit is work_mem.


Well, I'm pretty sure that having more work_mem is a good thing (tm) normally ;-)

Well, that's what I always thought too! :-) 
 

So, to sum this up (and make someone more competent bite on it maybe ;-), on your SMP, FSB, "fake-multicore" system all "hash"-related works that potentially switch to different implementations internally (but w/out telling us so) when given more work_mem are slower.

I'm pretty sure you're hitting some subtle, memory-access-related cornercase here.

The L2 cache of your X7350 CPUs is 2MB, could you run the tests with, say, 1, 2, 4 and 8MB of work_mem and post the results?

-- 
Gunnar "Nick" Bluth
RHCE/SCLA

Mobil   +49 172 8853339
Email: gunnar.bluth@pro-open.de
__________________________________________________________________________
In 1984 mainstream users were choosing VMS over UNIX.  Ten years later
they are choosing Windows over UNIX.  What part of that message aren't you
getting? - Tom Payne

В списке pgsql-performance по дате отправления:

Предыдущее
От: Jeff Janes
Дата:
Сообщение: Re: How to keep queries low latency as concurrency increases
Следующее
От: Claudio Freire
Дата:
Сообщение: Re: Constraint exclusion in views