Re: [HACKERS] Effect of changing the value for PARALLEL_TUPLE_QUEUE_SIZE

Поиск
Список
Период
Сортировка
От Robert Haas
Тема Re: [HACKERS] Effect of changing the value for PARALLEL_TUPLE_QUEUE_SIZE
Дата
Msg-id CA+TgmoYCCp7_NUVOzKqA9hR-nz4YWZpkkVPJPs1a=BDg8tvLrQ@mail.gmail.com
обсуждение исходный текст
Ответ на Re: [HACKERS] Effect of changing the value for PARALLEL_TUPLE_QUEUE_SIZE  (Ashutosh Bapat <ashutosh.bapat@enterprisedb.com>)
Список pgsql-hackers
On Wed, May 31, 2017 at 2:35 AM, Ashutosh Bapat
<ashutosh.bapat@enterprisedb.com> wrote:
> AFAIK, work_mem comes from memory private to the process whereas this
> memory will come from the shared memory pool.

I don't think that really matters.  The point of limits like work_mem
is to avoid driving the machine into swap.  Allocating shared memory
might error out rather than causing swapping in some cases on some
systems, but any difference between private and shared memory is not
the real issue here.  The issue is overall memory consumption.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Robert Haas
Дата:
Сообщение: Re: [HACKERS] TAP backpatching policy
Следующее
От: Mark Dilger
Дата:
Сообщение: Re: [HACKERS] Use of non-restart-safe storage by temp_tablespaces