Re: effective_io_concurrency and NVMe devices

Поиск
Список
Период
Сортировка
Искать
От
Nathan Bossart
Тема
Re: effective_io_concurrency and NVMe devices
Дата
Msg-id
20220420175858.GA2634425@nathanxps13
Ответ на
Список
Дерево обсуждения
effective_io_concurrency and NVMe devices Bruce Momjian <bruce@momjian.us>
Re: effective_io_concurrency and NVMe devices Nathan Bossart <nathandbossart@gmail.com>
RE: effective_io_concurrency and NVMe devices Jakub Wartak <Jakub.Wartak@tomtom.com>
Re: effective_io_concurrency and NVMe devices David Rowley <dgrowleyml@gmail.com>
Re: effective_io_concurrency and NVMe devices Tomas Vondra <tomas.vondra@enterprisedb.com>
Re: effective_io_concurrency and NVMe devices Tomas Vondra <tomas.vondra@enterprisedb.com>
RE: effective_io_concurrency and NVMe devices Jakub Wartak <Jakub.Wartak@tomtom.com>
Re: effective_io_concurrency and NVMe devices Tomas Vondra <tomas.vondra@enterprisedb.com>
RE: effective_io_concurrency and NVMe devices Jakub Wartak <Jakub.Wartak@tomtom.com>
Re: effective_io_concurrency and NVMe devices Tomas Vondra <tomas.vondra@enterprisedb.com>
On Tue, Apr 19, 2022 at 10:56:05PM -0400, Bruce Momjian wrote:
> NVMe devices have a maximum queue length of 64k:
> 
> 	https://blog.westerndigital.com/nvme-queues-explained/
> 
> but our effective_io_concurrency maximum is 1,000:
> 
> 	test=> set effective_io_concurrency = 1001;
> 	ERROR:  1001 is outside the valid range for parameter "effective_io_concurrency" (0 .. 1000)
> 
> Should we increase its maximum to 64k?  Backpatched?  (SATA has a
> maximum queue length of 256.)

If there are demonstrable improvements with higher values, this seems
reasonable to me.  I would even suggest removing the limit completely so
this doesn't need to be revisited in the future.

-- 
Nathan Bossart
Amazon Web Services: https://aws.amazon.com


В списке pgsql-hackers по дате отправления
От: Nathan Bossart
Дата:
От: Tomas Vondra
Дата:
FAQ