[36969] in North American Network Operators' Group
RE: jumbo frames
daemon@ATHENA.MIT.EDU (Roeland Meyer)
Fri Apr 27 12:25:36 2001
Message-ID: <9DC8BBAD4FF100408FC7D18D1F0922860E4544@condor.mhsc.com>
From: Roeland Meyer <rmeyer@mhsc.com>
To: "'Rowland, Alan D'" <alan_r1@corp.earthlink.net>,
'Kurt Kayser' <kurt@noris.de>, Tony Hain <alh-ietf@tndh.net>
Cc: Roeland Meyer <rmeyer@mhsc.com>,
John Fraizer <nanog@Overkill.EnterZone.Net>,
Paul Lantinga <prl@q9.com>, nanog@merit.edu
Date: Fri, 27 Apr 2001 09:11:58 -0700
MIME-Version: 1.0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Errors-To: owner-nanog-outgoing@merit.edu
In both cases, it is a bandwidth issue. You either consume the =
bandwidth in
the CPU or consume it on the wire. CPU is cheaper and often =
under-utilized.
Of the two, CPU bandwidth is also cheaper/easier to upgrade.
> -----Original Message-----
> From: Rowland, Alan D [mailto:alan_r1@corp.earthlink.net]
> Sent: Friday, April 27, 2001 8:45 AM
> To: 'Kurt Kayser'; Tony Hain
> Cc: Roeland Meyer; John Fraizer; Paul Lantinga; nanog@merit.edu
> Subject: RE: jumbo frames
>=20
>=20
> I am not an EE but maybe if you rephrased the question as
>=20
> Which is greater, the cpu cycles to assemble/dissemble jumbo=20
> frames or the
> additional cycles/bandwidth of more numerous ACK packets?
>=20
> Then again, I may be way out of my depth here.
>=20
> -Al
>=20
> -----Original Message-----
> From: Kurt Kayser [mailto:kurt@noris.de]
> Sent: Friday, April 27, 2001 8:07 AM
> To: Tony Hain
> Cc: Roeland Meyer; John Fraizer; Paul Lantinga; nanog@merit.edu
> Subject: Re: jumbo frames
>=20
>=20
>=20
> Hi,
>=20
> Isn't it a lot more cpu-intensive to 'collect' some 1500-byte frames
> into some larger bucket, reassemble it into a jumbo-frame=20
> when the next
> box has to chop it in order to send it out on a Sonet/PPP/etc=20
> interface
> which=20
> might have a smaller MTU again?
>=20
> Doesn't make too much sense to me. I guess that was Tony's=20
> aim as well..
>=20
> Kurt
> =20
> > Roeland you are talking about jumbo frames from the end=20
> system lan, while
> > John is talking about only using the jumbo frames between=20
> the routers. My
> > point was that in John's environment the packets will all=20
> be 1500 since
> the
> > packets are restricted to that size just to get to the=20
> router with the GE
> > interface. I understand that there are perf gains as long=20
> as the entire
> path
> > supports the larger packets, but I don't understand the=20
> claim that having
> a
> > bigger pipe in the middle helps.
> >=20
> > Tony
> >=20
> --=20
> noris network AG * tel +49 911 93 52-0 * internet
> Kilianstra=DFe 142 * fax +49 911 93 52-100 * solution
> 90425 N=FCrnberg * http://www.noris.net * provider
>=20
