Re: random_page_costs - are defaults of 4.0 realistic for SCSI RAID 1

Поиск
Список
Период
Сортировка
От Mark Mielke
Тема Re: random_page_costs - are defaults of 4.0 realistic for SCSI RAID 1
Дата
Msg-id 46E5AEA8.4010809@mark.mielke.cc
обсуждение исходный текст
Ответ на Re: random_page_costs - are defaults of 4.0 realistic for SCSI RAID 1  ("Luke Lonergan" <llonergan@greenplum.com>)
Список pgsql-performance
Luke Lonergan wrote:
> Should be a lot higher, something like 10-15 is approximating accurate.
>
In my own case, I have a much smaller database that I normally work
with, where everything should fit in memory (100 Mbytes?), and reducing
it to 3.0 has resulted in consistently better timings for me. I think
this means that the planner doesn't understand my database size :
effective memory size ratio. :-)

Anyways - my point is that if you change the default to 10 you may hurt
people like me.

Cheers,
mark

--
Mark Mielke <mark@mielke.cc>

В списке pgsql-performance по дате отправления:

Предыдущее
От: "Luke Lonergan"
Дата:
Сообщение: Re: random_page_costs - are defaults of 4.0 realistic for SCSI RAID 1
Следующее
От: Josh Berkus
Дата:
Сообщение: Re: random_page_costs - are defaults of 4.0 realistic for SCSI RAID 1