Date: Mon, 4 Aug 2014 09:53:50 -0700 From: David Wolfskill <david@catwhisker.org> To: Stefan Parvu <sparvu@systemdatarecorder.org> Cc: freebsd-hackers@freebsd.org Subject: Re: cpu utilization break down via sysctl Message-ID: <20140804165350.GQ1228@albert.catwhisker.org> In-Reply-To: <20140804194036.be325ee5822bbcb2623db358@systemdatarecorder.org> References: <20140804190923.95fec6a00d175ffee89c4af5@systemdatarecorder.org> <20140804162056.GP1228@albert.catwhisker.org> <20140804194036.be325ee5822bbcb2623db358@systemdatarecorder.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--5KxTQ9fdN6Op3ksq
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Mon, Aug 04, 2014 at 07:40:36PM +0300, Stefan Parvu wrote:
> > Sure; I've done the data acquisition with sysctl & awk, then
> > post-processed that with a bit of Perl. I wrote some code in R to
> > generate graphs.... I did some similar stuff using RRDTool a whlie
> > back.
>=20
> Im planning in using BSD::Sysctl to fetch all needed values.
For my data acquisition, I wasn't assuming that Perl would be available
-- I therefore intentionally restricted myself to tools in base
FreeBSD. Your situation is, of course, almost certainly different. :-)
> ...
> I see. Cool. If I understood right these will be the break down for 4 vcp=
u system:
>=20
> $ sysctl kern.cp_times
> kern.cp_times: 28845 2060 22253 1805 1197920 27677 1477 22399 2109 119918=
0 32342 1649 19499 894 1198458 29670 1678 20324 1471 1199699
>=20
> which maps to:
>=20
> time1: cpu0: user_28845: nice_2060: system_22253: intr_1805: idle_1197920
> time1: cpu1: user_27677: nice_1477 system_22399 intr_2109: idle_1199180=
=20
> time1: cpu2: user_32342: nice_1649: system_19499: intr_894: idle_1198458
> time1: cpu3 user_29670: nice_1678: system_20324: intr_1471: idle_1199699
Err... Well... Maybe? The numeric values are the values of the counters
at the time the information was acquired.
I'd probably depict them more as:
user nice system intr idle
cpu0: 28845 2060 22253 1805 1197920
cpu1: 27677 1477 22399 2109 1199180
cpu2: 32342 1649 19499 894 1198458
cpu3: 29670 1678 20324 1471 1199699
> ..
> Right I need to do delta between 2 readings to get the correct values.
Yes.
> Same as in KSTAT in Solaris perl interface.=20
It's been a long time since I did anything with Solaris. :-}
> One more check here:=20
>=20
> CPU_Utilization =3D User + System + Nice + Intr ?=20
If you mean total non-idle CPU, yes. For determining available CPU, you
need to include idle.
> Many thanks indeed. I need to experiment with these ...
Glad to help; the source for top(1) can also be instructive.
Peace,
david
--=20
David H. Wolfskill david@catwhisker.org
Taliban: Evil cowards with guns afraid of truth from a 14-year old girl.
See http://www.catwhisker.org/~david/publickey.gpg for my public key.
--5KxTQ9fdN6Op3ksq
Content-Type: application/pgp-signature
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=2Ifl
-----END PGP SIGNATURE-----
--5KxTQ9fdN6Op3ksq--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20140804165350.GQ1228>