[ee122] hmwk problem Number 2.16
Jorge Ortiz
jortiz at cs.berkeley.edu
Thu Sep 20 22:12:56 PDT 2007
MTU refers to the whole size of the packet (with all the headers that
frame the data -- i explain the difference in the next paragraph). So
for a datagram packet, this includes both the IP header AND the
Ethernet header (or whatever link-layer header your putting on the
packet).
The distinction between packet and frame is a bit subtle. A packet is
essentially encased by a frame. So when you append the header to the
front of the packet (and the back of the packet if there is trailer
information), you are framing the packet inside a frame of the type of
header you're putting on the packet.
So for example, you place an IP packet inside an Ethernet frame when
you send it over Ethernet.
jorge
On 9/20/07, Nescio Nomen <nescionomen at gmail.com> wrote:
> The definition I'm using is on P. 242-243. I'm not sure there is a
> conflict. You're interpreting "data" in your sentence to mean "payload,"
> I'm not sure this is the case. Assuming the book does not contradict
> itself, looking at the definition on P.242-243, I would guess that "data" in
> your sentence means "the stuff in the link layer frame [including the
> header]." But it's not very precise language so maybe I'm wrong and the
> book does contradict itself.
>
>
> On 9/20/07, Jonathan D. Ellithorpe <jde at berkeley.edu> wrote:
> > Ah very interesting. That is why you and I are saying different things.
> >
> > On page 339 of the textbook it says:
> >
> > "The maximum amount of data that a link layer frame can carry is called
> > the maximum transmission unit (MTU)."
> >
> > Which conflicts directly with:
> >
> > "MTU is the length of the largest link-layer frame that can be sent by
> > the local sending host"
> >
> >
> >
> > Nescio Nomen wrote:
> > > Is it defined for packets or for frames? I'm finding conflicting
> > > definitions on google. The book says that the MTU is the "length of
> > > the largest link-layer frame that can be sent by the local sending
> > > host"-- hence my confusion since we're not provided any link layer
> > > header info. But, if I understand what you've said earlier, we're to
> > > assume it's strictly going to only have IP header anyway.
> > >
> > > On 9/20/07, *Jorge Ortiz* < jortiz at cs.berkeley.edu
> > > <mailto:jortiz at cs.berkeley.edu>> wrote:
> > >
> > > The MTU is the largest sized packet that can be passed. This
> includes
> > > header information.
> > >
> > >
> > > On 9/20/07, Nescio Nomen <nescionomen at gmail.com
> > > <mailto:nescionomen at gmail.com >> wrote:
> > > > Oh wait, I think I understand-- you're saying that MTU does not
> > > include the
> > > > link header size?
> > > >
> > > >
> > > > On 9/20/07, Nescio Nomen < nescionomen at gmail.com
> > > <mailto:nescionomen at gmail.com>> wrote:
> > > > > I'm not sure I understand. (in a real life problem) why
> > > wouldn't we need
> > > > to know the size of the frame header just because we are given
> > > the MTU?
> > > > Wouldn't we try to find the amount of 'real data' that could fit
> > > by doing
> > > > something along the lines of MTU - frame header size - IP header
> > > size -
> > > > Layer 4 header size?
> > > > >
> > > > >
> > > > >
> > > > > On 9/20/07, Jonathan D. Ellithorpe < jde at berkeley.edu
> > > <mailto:jde at berkeley.edu>> wrote:
> > > > > > I think the MTU is actually defined as being the maximum
> > > amount of data
> > > > > > that a link layer frame can encapsulate. Thus, we don't need
> > > to consider
> > > > > > the size of the frame, since we're just given the MTU of the
> > > link-layer.
> > > > > >
> > > > > >
> > > > > > Jorge Ortiz wrote:
> > > > > > > On 9/20/07, Nescio Nomen < nescionomen at gmail.com
> > > <mailto:nescionomen at gmail.com>> wrote:
> > > > > > >
> > > > > > >> For hmwk Problem #2, P16, do we know if the datagram will
> > > have a TCP
> > > > header
> > > > > > >> inside? (Is it necessarily the case that a datagram always
> > > > encapsulated a
> > > > > > >> TCP header?)
> > > > > > >>
> > > > > > >
> > > > > > > If it's specified as only being a plane datagram (as it is
> > > in the
> > > > > > > problem), we only need to include the IP header.
> > > > > > >
> > > > > > >
> > > > > > >> Also, when we are transmitting into the link, how big is
> the
> > > > > > >> frame header?
> > > > > > >>
> > > > > > >
> > > > > > > Treat the packet as having only an IP header and the data
> > > you wish to
> > > > send.
> > > > > > >
> > > > > > >
> > > > > > >> We haven't really talked about this layer yet. Does the
> > > size
> > > > > > >> of the frame header depend on the technology, i.e. optical?
> > > > > > >>
> > > > > > >
> > > > > > > It may, yes. Different mediums may have different header
> > > definitions.
> > > > > > >
> > > > > > > jorge
> > > > > > >
> > > > > > >
> > > > > > >>
> _______________________________________________
> > > > > > >> ee122 mailing list
> > > > > > >> ee122 at mailman.ICSI.Berkeley.EDU
> > > <mailto:ee122 at mailman.ICSI.Berkeley.EDU>
> > > > > > >>
> > > >
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/ee122
> > > <
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/ee122>
> > > > > > >>
> > > > > > >>
> > > > > > >>
> > > > > > >
> _______________________________________________
> > > > > > > ee122 mailing list
> > > > > > > ee122 at mailman.ICSI.Berkeley.EDU
> > > <mailto: ee122 at mailman.ICSI.Berkeley.EDU>
> > > > > > >
> > > >
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/ee122
> > > <
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/ee122>
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> _______________________________________________
> > > > > > ee122 mailing list
> > > > > > ee122 at mailman.ICSI.Berkeley.EDU
> > > <mailto:ee122 at mailman.ICSI.Berkeley.EDU>
> > > > > >
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/ee122
> > > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > > _______________________________________________
> > > > ee122 mailing list
> > > > ee122 at mailman.ICSI.Berkeley.EDU
> > > <mailto:ee122 at mailman.ICSI.Berkeley.EDU>
> > > >
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/ee122
> > > >
> > > >
> > >
> > >
> >
> > _______________________________________________
> > ee122 mailing list
> > ee122 at mailman.ICSI.Berkeley.EDU
> > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/ee122
> >
>
>
> _______________________________________________
> ee122 mailing list
> ee122 at mailman.ICSI.Berkeley.EDU
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/ee122
>
>
More information about the ee122
mailing list