Re: Remove the comment on the countereffectiveness of large shared_buffers on Windows

Поиск
Список
Период
Сортировка
От Jeff Janes
Тема Re: Remove the comment on the countereffectiveness of large shared_buffers on Windows
Дата
Msg-id CAMkU=1yQGhW8Q_x22bSpVUUkVEpYGNZN+Qx6sEE3XokCoJuLmw@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Remove the comment on the countereffectiveness of large shared_buffers on Windows  (Amit Kapila <amit.kapila16@gmail.com>)
Ответы Re: Remove the comment on the countereffectiveness of large shared_buffers on Windows
Список pgsql-hackers
Дерево обсуждения
On Mon, Nov 7, 2016 at 7:03 PM, Amit Kapila <amit.kapila16@gmail.com> wrote:
On Tue, Nov 8, 2016 at 5:12 AM, Jeff Janes <jeff.janes@gmail.com> wrote:
> On Mon, Nov 7, 2016 at 5:55 AM, Amit Kapila <amit.kapila16@gmail.com> wrote:
>>
>>
>> Isn't it somewhat strange that writes are showing big win whereas
>> reads doesn't show much win?
>
>
> I don't find that unusual, and have seen the same thing on Linux.
>
> With small shared_buffers, you are constantly throwing dirty buffers at the
> kernel in no particular order, and the kernel does a poor job of predicting
> when the same buffer will be dirtied repeatedly and only needs the final
> version of the data actually written to disk.
>

Okay and I think partially it might be because we don't have writeback
optimization (done in 9.6) for Windows. 

Is the writeback optimization the introduction of checkpoint_flush_after, or is it something else? 

If it is checkpoint_flush_after, then I don't think that that is related.  In fact, they operate in opposite directions.  The writeback optimization forces the kernel to be more eager about writing out dirty data, so it doesn't have a giant pile of it when the fsyncs comes at the end of the checkpoint.  Using a large shared_buffers forces it to be less eager, by not turning the dirty buffers over to the kernel as often.
 
However, still the broader
question stands that whether above data is sufficient to say that we
can recommend the settings of shared_buffers on Windows similar to
Linux?

We do have evidence that the old advice is out of date.  So I think we should just remove it.  We have no evidence that the optimal Windows setting on modern Windows is the same as it is for Linux, but also no evidence that it is not the same.
 
Cheers,

Jeff

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tomas Vondra
Дата:
Сообщение: Re: xlogreader.c fails with FATAL on a cluster with 4kB block size
Следующее
От: Pavel Stehule
Дата:
Сообщение: Re: proposal: psql \setfileref