Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 10 Feb 2014 11:07:36 -0500
From:      Glen Barber <gjb@FreeBSD.org>
To:        Tim Daneliuk <tundra@tundraware.com>
Cc:        freebsd-stable@freebsd.org
Subject:   Re: And Here I Thought buildworld/makeworld Was IO Bound
Message-ID:  <20140210160736.GC1629@glenbarber.us>
In-Reply-To: <52F8F426.9000003@tundraware.com>
References:  <52F84AF8.8050007@tundraware.com> <52F86768.9000109@googlemail.com> <52F8E8C6.2060909@tundraware.com> <20140210152407.GB1629@glenbarber.us> <52F8F426.9000003@tundraware.com>

next in thread | previous in thread | raw e-mail | index | archive | help

--qXCixuLMVvZDruUh
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Mon, Feb 10, 2014 at 09:45:42AM -0600, Tim Daneliuk wrote:
> >>Well ... it has to hit the disk sooner or later.  But, if the
> >>frequency of physical writes is low because of aggregated
> >>IO from the cache, I guess that would tend to make the whole
> >>business more CPU bound than IO bound.  I just found this surprising.
> >>
> >
> >What is the underlying filesystem?
> >
>=20
> UFS w/Softupdates, no journaling.
>=20
>=20
> [...]

> >>Yeah, I wonder what other people are seeing for a full buildworld/kerne=
l and/or what
> >>the master machines at FreeBSD.org do in this regards.
> >>
> >>Would anyone else care to share with the class?
> >>
> >
> >If you mean "what do the machines do with regard to tuning", the only
> >specific tuning is turning off atime.
> >
>=20
>=20
> Could you comment a bit more about this please?  How you do it,
> rationale', etc.
>=20

Each build happens in its own ZFS dataset, so atime is disabled during
dataset creation (zfs create -o atime=3Doff zroot/11-amd64-GENERIC-snap).

The tuning(7) manual has more details, but basically each time a file is
read, a separate write happens to update the file access time.  Turning
off atime eliminates these excessive writes.

> >If you mean "what do the machines do as far as overall build time", with
> >a clean obj/ directory, 35 minutes sounds about right.  I do not have
> >exact numbers.
> >
> >The machine that currently produces weekly snapshot images uses '-j10'
> >for buildworld and '-j6' for buildkernel for each build.  It also runs
> >the builds in parallel.  Even with three parallel 'make -j10 buildworld'
> >(one for head/, stable/10/, and stable/9/), disk IO is minimal.  The
> >only time the disk IO becomes a bottleneck is when creating the
> >distribution files (base.txz, kernel.txz, etc.).
>=20
>=20
> What sort of CPU/Mem/Disk is that machine?
>=20

This machine is 48-core (4 x 12 cores) Opteron 6174 (2.2GHz), 128GB RAM,
with 5 drives in a raidz1 on a PERC H700 controller.

Glen


--qXCixuLMVvZDruUh
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (FreeBSD)
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=Wm6a
-----END PGP SIGNATURE-----

--qXCixuLMVvZDruUh--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20140210160736.GC1629>