Re: high user cpu, massive SELECTs, no io waiting problem - Mailing list pgsql-performance

From Scott Marlowe
Subject Re: high user cpu, massive SELECTs, no io waiting problem
Date
Msg-id AANLkTikb3XJih-O+fJtOsWWAtkq8FG9iCCR8N-_uMXyp@mail.gmail.com
Whole thread Raw
In response to Re: high user cpu, massive SELECTs, no io waiting problem  (Marti Raudsepp <marti@juffo.org>)
Responses Re: high user cpu, massive SELECTs, no io waiting problem  ("Strange, John W" <john.w.strange@jpmchase.com>)
List pgsql-performance
On Wed, Feb 16, 2011 at 6:44 AM, Marti Raudsepp <marti@juffo.org> wrote:
> On Tue, Feb 15, 2011 at 20:01, Scott Marlowe <scott.marlowe@gmail.com> wrote:
>> run htop and look for red.  if youi've got lots of red bar on each CPU
>> but no io wait then it's waiting for memory access.
>
> I don't think this is true. AFAICT the red bar refers to "system
> time", time that's spent in the kernel -- either in syscalls or kernel
> background threads.

My point being that if you've got a lot of RED it'll be the OS waiting
for memory access.  Trust me, when we start to hit our memory
bandwidth (in the 70 to 80 GB/s range) we start to get more and more
red and more and more kernel wait time.

pgsql-performance by date:

Previous
From: Greg Smith
Date:
Subject: Re: high user cpu, massive SELECTs, no io waiting problem
Next
From: Scott Marlowe
Date:
Subject: Re: high user cpu, massive SELECTs, no io waiting problem