Re: ThisTimeLineID in checkpointer and bgwriter processes - Mailing list pgsql-hackers

From Andres Freund
Subject Re: ThisTimeLineID in checkpointer and bgwriter processes
Date
Msg-id 20121220133920.GD4303@awork2.anarazel.de
Whole thread Raw
In response to Re: ThisTimeLineID in checkpointer and bgwriter processes  (Simon Riggs <simon@2ndQuadrant.com>)
List pgsql-hackers
On 2012-12-20 13:32:36 +0000, Simon Riggs wrote:
> On 20 December 2012 13:19, Amit Kapila <amit.kapila@huawei.com> wrote:
>
> >> So I think we're good on that front. But I'll add a comment there, and
> >> use 0 explicitly instead of ThisTimeLineID, for clarity.
> >
> > True, it might not have any functionality effect in RemoveOldXlogFiles().
> > However it can be used in PreallocXlogFiles()->XLogFileInit() as well.
>
> PreallocXlogFiles() should have been put to the sword long ago. It's a
> performance tweak aimed at people without a performance problem in
> this area.

Creating, zeroing & fsync()'ing a 16MB file shouldn't be done in
individual transactions because it would increase latency noticeably. So
it seems it addresses a real performance problem. What do you dislike
about it?

Greetings,

Andres Freund

--Andres Freund                       http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services



pgsql-hackers by date:

Previous
From: Simon Riggs
Date:
Subject: Re: ThisTimeLineID in checkpointer and bgwriter processes
Next
From: Peter Eisentraut
Date:
Subject: Re: Parser Cruft in gram.y