From: | Johnny Tan <johnnydtan(at)gmail(dot)com> |
---|---|
To: | David Whittaker <dave(at)iradix(dot)com> |
Cc: | Josh Krupka <jkrupka(at)gmail(dot)com>, Alex Kahn <alex(at)paperlesspost(dot)com>, PostgreSQL Performance <pgsql-performance(at)postgresql(dot)org> |
Subject: | Re: postgresql.conf recommendations |
Date: | 2013-02-06 19:45:38 |
Message-ID: | CABMVzL2GYAvWkBC=ik2s=D5=WT1qFWi9vVcsZpZSozSFLs1fng@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
On Wed, Feb 6, 2013 at 2:13 PM, David Whittaker <dave(at)iradix(dot)com> wrote:
> We disabled THP all together, with the thought that we might re-enable
> without defrag if we got positive results. At this point I don't think THP
> is the root cause though, so I'm curious to see if anyone else gets
> positive results from disabling it. We definitely haven't seen any
> performance hit from turning it off.
>
We are considering disabling THP, although we aren't seeing the errors in
syslog that David mentioned.
Josh: What made you think of THP? I'm curious if there's a discussion you
can point me to. Since you mentioned it, I've been looking more into it,
and there isn't too much. In fact, this post makes it sound like enabling
it fixes a similar problem to what we're seeing -- i.e., %system shoots up
during the spikes:
http://www.pythian.com/blog/performance-tuning-hugepages-in-linux/
johnny
From | Date | Subject | |
---|---|---|---|
Next Message | Johnny Tan | 2013-02-06 19:47:54 | Re: postgresql.conf recommendations |
Previous Message | Johnny Tan | 2013-02-06 19:33:45 | Re: postgresql.conf recommendations |