Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 1 Jun 2004 15:59:23 -0700
From:      Brooks Davis <brooks@one-eyed-alien.net>
To:        Luke <luked@pobox.com>
Cc:        freebsd-current@freebsd.org
Subject:   Re: Cap on network speed in CURRENT?
Message-ID:  <20040601225922.GA20044@Odin.AC.HMC.Edu>
In-Reply-To: <Pine.NEB.4.60.0406011524390.20463@otaku.freeshell.org>
References:  <Pine.NEB.4.60.0406011524390.20463@otaku.freeshell.org>

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

--C7zPtVaVf+AK4Oqc
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Jun 01, 2004 at 03:38:18PM -0700, Luke wrote:
>=20
> I've got a 100Mbps LAN with ethernet cards that should be capable of usin=
g=20
> it, yet the highest transfer rates I seem to be able to get out of my=20
> FreeBSD box are 260KB/s receiving and 341KB/s sending with around 200KB/s=
=20
> being more normal.
>=20
> I realize that there are hundreds of factors that could be influencing=20
> this, but I came across this recent article that made me wonder if this i=
s=20
> some kind of hardcoded limit:
> http://www.freebsd.org/news/status/report-jan-2004-feb-2004.html#Automati=
c-sizing-of-TCP-send-buffers
>=20
> Is this article saying that my network speed is limited by a small=20
> static TCP buffer size?  If so, is there some way that I can increase tha=
t=20
> buffer size to improve performance?  The primary function of this machine=
=20
> is to move large amounts of data across my network, so I'm willing to=20
> experiment with increasing the buffer size if it's not too difficult.

On a LAN, buffer size has minimal effect except at very high speeds.
Without tuning, two 5.x boxes with gigabit interfaces connected to a
Cisco 6513 switch (one 5/10/04 and one 2/20/04) reached 187Mbps in
iperf.  You're problems symptoms sound like duplex mismatch or bad
hardware to me.

-- Brooks

--=20
Any statement of the form "X is the one, true Y" is FALSE.
PGP fingerprint 655D 519C 26A7 82E7 2529  9BF0 5D8E 8BE9 F238 1AD4

--C7zPtVaVf+AK4Oqc
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQFAvQpKXY6L6fI4GtQRAg8uAKDBCZpB6c5rjs7bEtb2irpILJcS5ACfVeLD
/bh/RE7DKj6VkSHsOHm3p4o=
=RVXA
-----END PGP SIGNATURE-----

--C7zPtVaVf+AK4Oqc--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040601225922.GA20044>