Re: How filesystems matter with PostgreSQL

От: Jon Schewe
Тема: Re: How filesystems matter with PostgreSQL
Дата: ,
Msg-id: 4C0AD280.5020105@mtu.net
(см: обсуждение, исходный текст)
Ответ на: Re: How filesystems matter with PostgreSQL  (Greg Smith)
Ответы: Re: How filesystems matter with PostgreSQL  (Greg Smith)
Список: pgsql-performance

Скрыть дерево обсуждения

How filesystems matter with PostgreSQL  (Jon Schewe, )
 Re: How filesystems matter with PostgreSQL  ("J. Roeleveld", )
 Re: How filesystems matter with PostgreSQL  (Andres Freund, )
  Re: How filesystems matter with PostgreSQL  (Tom Lane, )
   Re: How filesystems matter with PostgreSQL  (Andres Freund, )
    Re: How filesystems matter with PostgreSQL  (Bryan Hinton, )
     Re: How filesystems matter with PostgreSQL  (Jon Schewe, )
    Re: How filesystems matter with PostgreSQL  (Jon Schewe, )
     Re: How filesystems matter with PostgreSQL  (Bryan Hinton, )
      Re: How filesystems matter with PostgreSQL  (Jon Schewe, )
     Re: How filesystems matter with PostgreSQL  (Andres Freund, )
      Re: How filesystems matter with PostgreSQL  (Jon Schewe, )
 Re: How filesystems matter with PostgreSQL  (Greg Smith, )
  Re: How filesystems matter with PostgreSQL  (Jon Schewe, )
   Re: How filesystems matter with PostgreSQL  (Greg Smith, )
    Re: How filesystems matter with PostgreSQL  (Jon Schewe, )
     Re: How filesystems matter with PostgreSQL  (Scott Marlowe, )
      Re: How filesystems matter with PostgreSQL  (Jon Schewe, )
       Re: How filesystems matter with PostgreSQL  (Scott Marlowe, )
        Re: How filesystems matter with PostgreSQL  (Jon Schewe, )
         Re: How filesystems matter with PostgreSQL  (Scott Marlowe, )
     Re: How filesystems matter with PostgreSQL  (Ron Mayer, )
      Re: How filesystems matter with PostgreSQL  (Craig Ringer, )

On 06/05/2010 05:36 PM, Greg Smith wrote:
> Jon Schewe wrote:
>> The tests were all done on an opensuse 11.2 64-bit machine,
>> on the same hard drive (just ran mkfs between each test) on the same
>> input with the same code base.
>
> So no controller card, just the motherboard and a single hard drive?
Correct.
>   If that's the case, what you've measured is which filesystems are
> safe because they default to flushing drive cache (the ones that take
> around 15 minutes) and which do not (the ones that take >=around 2
> hours).  You can't make ext3 flush the cache correctly no matter what
> you do with barriers, they just don't work on ext3 the way PostgreSQL
> needs them to.
>
So the 15 minute runs are doing it correctly and safely, but the slow
ones are doing the wrong thing? That would imply that ext3 is the safe
one. But your last statement suggests that ext3 is doing the wrong thing.


В списке pgsql-performance по дате сообщения:

От: Scott Marlowe
Дата:
Сообщение: Re: How filesystems matter with PostgreSQL
От: Ron Mayer
Дата:
Сообщение: Re: How filesystems matter with PostgreSQL