Re: CPU load spikes when CentOS tries to reclaim 'cached' memory

From: Merlin Moncure <mmoncure(at)gmail(dot)com>
To: Jeff Janes <jeff(dot)janes(at)gmail(dot)com>
Cc: vlasmarias <vlasmarias(at)contigo(dot)com>, "pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org>
Subject: Re: CPU load spikes when CentOS tries to reclaim 'cached' memory
Date: 2014-06-05 19:23:46
Message-ID: CAHyXU0x+FJZb+J6VHQpndSiE_7y88OLY1Jd_4TKb-pUqHm-Tuw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Thu, Jun 5, 2014 at 10:58 AM, Jeff Janes <jeff(dot)janes(at)gmail(dot)com> wrote:
> This sounds like a kernel problem, probably either the zone reclaim issue,
> or the transparent huge pages issue.

I at first thought maybe same, but I don't think THP was introduced
until 2.6.38...OP is running 2.6.32-431.11.2.el6.x86_6. Maybe it's
NUMA related, but would not be idiomatic of NUMA issues as I
understand them (poor memory utilization/high IO utilization). Would
be a very cheap/easy thing to try though.

Is this server virtualized?

merlin

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Weinzierl Stefan 2014-06-05 19:36:08 Seqscan on big table, when an Index-Usage should be possible
Previous Message Jeff Janes 2014-06-05 15:58:29 Re: CPU load spikes when CentOS tries to reclaim 'cached' memory