linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrea Righi <andrea@betterlinux.com>
To: Paul Menage <paul@paulmenage.org>, Ingo Molnar <mingo@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org
Subject: [PATCH RFC 0/3] per cpuset load average
Date: Thu,  4 Oct 2012 01:05:09 +0200	[thread overview]
Message-ID: <1349305512-3428-1-git-send-email-andrea@betterlinux.com> (raw)

[ Premise: I'm not considering this patch suitable for inclusion, mostly
because I'm not convinced at all of the interface and I don't even know if
cpuset is the right place for this feature. However, I just wrote it for a
project and I would like to improve it, better integrate it in the kernel and
share the code in general, in case someone else is also interested to use this
feature. ]

Overview
~~~~~~~~
The cpusets subsystem allows to assign a different set of CPUs to a cgroup. A
typical use case is to split large systems in small CPU/memory partitions and
isolate certain users/applications in these subsets of the system.

Sometimes, to have a quick overview of each partition's state, we may be
interested to get the load average of the CPUs assigned to a particular cpuset,
rather than the global load average of the system.

Proposed solution
~~~~~~~~~~~~~~~~~
The proposal is to add a new file in the cpuset subsystem to report the load
average of the CPUs assinged to a particular cpuset cgroup.

Example:

 # echo 0-1 > /sys/fs/cgroup/cpuset/foo/cpuset.cpus
 # echo 2-3 > /sys/fs/cgroup/cpuset/bar/cpuset.cpus

 # echo $$ > /sys/fs/cgroup/cpuset/foo/tasks
 # for i in `seq 4`; do yes > /dev/null & done

 ... after ~5mins ...

 # cat /proc/loadavg /sys/fs/cgroup/cpuset/{foo,bar}/cpuset.loadavg
 3.99 2.66 1.24 6/377 2855
 3.98 2.64 1.20
 0.01 0.02 0.04

In this case we can easily find that the cpuset "foo" is the most busy in the
system.

[PATCH RFC 1/3] sched: introduce distinct per-cpu load average
[PATCH RFC 2/3] cpusets: add load avgerage interface
[PATCH RFC 3/3] cpusets: add documentation of the loadavg file

 include/linux/sched.h |    7 +++++
 kernel/cpuset.c       |   58 ++++++++++++++++++++++++++++++++++++
 kernel/sched/core.c   |   78 ++++++++++++++++++++++++++++++++++++++++++++++---
 3 files changed, 139 insertions(+), 4 deletions(-)

             reply	other threads:[~2012-10-03 23:07 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-03 23:05 Andrea Righi [this message]
2012-10-03 23:05 ` [PATCH RFC 1/3] sched: introduce distinct per-cpu load average Andrea Righi
2012-10-04  8:59   ` Peter Zijlstra
2012-10-04  9:43     ` Andrea Righi
2012-10-04 12:12       ` Peter Zijlstra
2012-10-04 17:19         ` Andrea Righi
2012-10-03 23:05 ` [PATCH RFC 2/3] cpusets: add load avgerage interface Andrea Righi
2012-10-03 23:05 ` [PATCH RFC 3/3] cpusets: add documentation of the loadavg file Andrea Righi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1349305512-3428-1-git-send-email-andrea@betterlinux.com \
    --to=andrea@betterlinux.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=paul@paulmenage.org \
    --cc=peterz@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).