pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
FreeBSD page size (was Re: The results of my PostgreSQL/filesystem performance tests)
2+ messages / 2 participants
[nested] [flat]

* FreeBSD page size (was Re: The results of my PostgreSQL/filesystem performance tests)
@ 2003-09-03 15:48  Vivek Khera <khera@kcilink.com>
  0 siblings, 1 reply; 2+ messages in thread

From: Vivek Khera @ 2003-09-03 15:48 UTC (permalink / raw)
  To: pgsql-performance

>>>>> "SC" == Sean Chittenden <sean@chittenden.org> writes:

>> I need to step in and do 2 things:
SC> Thanks for posting that.  Let me know if you have any questions while
SC> doing your testing.  I've found that using 16K blocks on FreeBSD
SC> results in about an 8% speedup in writes to the database, fwiw.

Just double checking: if I do this, then I need to halve the
parameters in postgresql.conf that involve buffers, specifically,
max_fsm_pages and shared_buffers.  I think max_fsm_pages should be
adjusted since the number of pages in the system overall has been
halved.

Anything else that should be re-tuned for this?

My tests are still running so I don't have numbers yet.

-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Vivek Khera, Ph.D.                Khera Communications, Inc.
Internet: khera@kciLink.com       Rockville, MD       +1-240-453-8497
AIM: vivekkhera Y!: vivek_khera   http://www.khera.org/~vivek/



^ permalink  raw  reply  [nested|flat] 2+ messages in thread

* Re: FreeBSD page size (was Re: The results of my
@ 2003-09-04 21:56  scott.marlowe <scott.marlowe@ihs.com>
  parent: Vivek Khera <khera@kcilink.com>
  0 siblings, 0 replies; 2+ messages in thread

From: scott.marlowe @ 2003-09-04 21:56 UTC (permalink / raw)
  To: Vivek Khera <khera@kcilink.com>; +Cc: pgsql-performance

On Wed, 3 Sep 2003, Vivek Khera wrote:

> >>>>> "SC" == Sean Chittenden <sean@chittenden.org> writes:
> 
> >> I need to step in and do 2 things:
> SC> Thanks for posting that.  Let me know if you have any questions while
> SC> doing your testing.  I've found that using 16K blocks on FreeBSD
> SC> results in about an 8% speedup in writes to the database, fwiw.
> 
> Just double checking: if I do this, then I need to halve the
> parameters in postgresql.conf that involve buffers, specifically,
> max_fsm_pages and shared_buffers.  I think max_fsm_pages should be
> adjusted since the number of pages in the system overall has been
> halved.
> 
> Anything else that should be re-tuned for this?

Yes, effective_cache_size as well is measured in pgsql blocks.




^ permalink  raw  reply  [nested|flat] 2+ messages in thread


end of thread, other threads:[~2003-09-04 21:56 UTC | newest]

Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2003-09-03 15:48 FreeBSD page size (was Re: The results of my PostgreSQL/filesystem performance tests) Vivek Khera <khera@kcilink.com>
2003-09-04 21:56 ` Re: FreeBSD page size (was Re: The results of my scott.marlowe <scott.marlowe@ihs.com>

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox