From: Jonathan Walsh <jwalsh@atl.lmco.com>
To: linux-kernel@vger.kernel.org
Cc: "Gautam H. Thaker" <gthaker@atl.lmco.com>, mingo@elte.hu
Subject: RE: ~5x greater CPU load for a networked application when using 2.6.15-rt15-smp vs. 2.6.12-1.1390_FC4
Date: Tue, 11 Jul 2006 14:08:32 -0400 [thread overview]
Message-ID: <44B3E920.9070907@atl.lmco.com> (raw)
As a follow up to previous emails (Gautam Thaker, Ingo Molnar, Ted Tso, et. al.) on the subject of large CPU overhead by the RT kernel when under heavy network load, I ran the following test in order to get more reasonable data. I have 19 nodes with 20 "virtual" node processes sending UDP messages to a single host at a rate of 100Hz for 38,000 packets per second. Using cyclesoak to determine cpu usage (over 240 samples, 1 sample per second), I found the following results:
RT kernel: linux-2.6.17-rt1-uni
Mean: 48.9%
Variance: 5.91
Standard Deviation: 2.43
Standard kernel: Standard Fedora Core 4
Mean: 23.2%
Variance: 0.237
Standard Deviation: 0.4867
Thus I found the average cpu load on the RT kernel to be 2.11 times that of the standard kernel. Hopefully this information will be of some use.
-Jonathan Walsh
Distributed Processing Lab; Lockheed Martin Adv. Tech. Labs
next reply other threads:[~2006-07-11 18:08 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-07-11 18:08 Jonathan Walsh [this message]
-- strict thread matches above, loose matches on Subject: below --
2006-02-23 19:55 ~5x greater CPU load for a networked application when using 2.6.15-rt15-smp vs. 2.6.12-1.1390_FC4 Gautam H Thaker
2006-02-23 20:15 ` Benjamin LaHaise
2006-02-23 20:58 ` Ingo Molnar
2006-02-23 21:06 ` Nish Aravamudan
2006-02-23 21:08 ` Ingo Molnar
2006-02-23 21:14 ` Nish Aravamudan
2006-02-23 22:07 ` Esben Nielsen
2006-02-24 8:03 ` Jan Engelhardt
2006-02-24 12:11 ` Andrew Morton
2006-02-24 20:06 ` Gautam H Thaker
2006-02-24 20:31 ` Andrew Morton
2006-02-24 20:44 ` Gautam H Thaker
2006-02-24 16:52 ` Theodore Ts'o
2006-02-24 19:25 ` Gautam H Thaker
2006-02-28 19:27 ` Matt Mackall
2006-02-28 22:19 ` Gautam H Thaker
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=44B3E920.9070907@atl.lmco.com \
--to=jwalsh@atl.lmco.com \
--cc=gthaker@atl.lmco.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/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.