All of lore.kernel.org
 help / color / mirror / Atom feed
From: Rob Gardner <rob.gardner@hp.com>
Cc: xen-devel <xen-devel@lists.sourceforge.net>
Subject: Re: Calculating real cpu usage of Xen domains	correctly! (PATCH)
Date: Sun, 27 Feb 2005 23:20:21 -0700	[thread overview]
Message-ID: <4222B825.8060106@hp.com> (raw)
In-Reply-To: <A95E2296287EAD4EB592B5DEEFCE0E9D1E32B8@liverpoolst.ad.cl.cam.ac.uk>

Ian Pratt wrote:
>>OK, here is a patch which provides fine grained cpu utilization 
>>reporting. Some notes:
> 
> 
> I really can't see the point of this patch -- it isn't necessary to log
> every scheduling operation to record CPU usage for a domain over
> multiple time scales. 

Yes, obviously you are correct. I am sorry for not explaining this patch 
better. Let me start from the beginning. We developed this code to 
collect various kinds of data. One of the things it does is measure cpu 
usage. However, that is not all it does, since as you rightly say, such 
complicated code is not necessary to to just measure cpu usage. When 
several people started talking about measuring cpu usage, I just 
volunteered this code to solve the problem. So I do not propose this 
code solely to measure cpu usage.

For instance, one thing the code can do is to measure how much time 
elapses while a domain is runnable, versus how much cpu time it actually 
receives. This metric might be one simple way of expressing 'quality of 
service.' In order to do this, some scheduler activity needs to be 
monitored. Hence the more heavy handed approach. We have other ideas in 
mind that require very fine grained measurements, some have been 
implemented, some not. I hope to share some of the other metrics soon.

Additionally, this is somewhat experimental code in that it was written 
this way in order to facilitate future measurements that we may not have 
  considered yet. Can we let some people try it out and see if it 
provides this framework?

Thanks,
Rob


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

  reply	other threads:[~2005-02-28  6:20 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-02-26 20:27 Calculating real cpu usage of Xen domains correctly! (PATCH) Ian Pratt
2005-02-28  6:20 ` Rob Gardner [this message]
  -- strict thread matches above, loose matches on Subject: below --
2005-02-24  0:48 Calculating real cpu usage of Xen domains correctly! Ian Pratt
2005-02-24  1:02 ` Anthony Liguori
2005-02-25  2:25   ` Matt Ayres
2005-02-25  2:38     ` Anthony Liguori
2005-02-25 16:44       ` Rob Gardner
2005-02-25 17:10         ` Anthony Liguori
2005-02-25 21:35           ` Calculating real cpu usage of Xen domains correctly! (PATCH) Rob Gardner

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=4222B825.8060106@hp.com \
    --to=rob.gardner@hp.com \
    --cc=xen-devel@lists.sourceforge.net \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.