Hi Jose,
comments inline.
On 19 Oct. 2012, at 18:04 , Jose Saldana <[email protected]
<mailto:[email protected]>> wrote:
Luigi,
What about this protocol stack for LISP within TCMTF?
(to be seen in Courier letters)
TCP/IP UDP/IP RTP/UDP/IP
\ | /
\ | / ------------------------------
\ | /
Nothing or ROHC or ECRTP or IPHC header compressing layer
|
| ------------------------------
|
PPPMUX or other mux protocols multiplexing layer
/ | \
/ | \ ------------------------------
/ LISP \
/ | \
GRE or L2TP UDP \ tunneling layer
| | MPLS
| | ------------------------------
IP IP
I don’t know if LISP could be included into TCMTF draft. What do you
think?
The graph looks good to me.
Concerning whether or not to include it in the draft, may be it is not
necessary to include details about LISP just mention that can be used.
The usage of LISP, because of its experimental nature compare to other
tunnelling options, can be documented in a different document, which, if
people are interested, I can (with your help) write.
Should it be defined inside the LISP WG?
I`ll let the LISP WG chairs answer on this point.
Nevertheless, IMHO, the LISP WG is chartered for the main spec. Other
usages of LISP can be discussed in other, more pertinent, WGs. Obviously
some form of collaboration will be necessary at least to make the LISP
WG aware of how other WGs plan to use LISP.
We have to talk about that LISP signaling which could be used in
TCMTF, as you say.
As I said before, these details can be documented elsewhere. If LISP is
mentioned in the main draft, then it can also be stated that the re-use
of the LISP signalling machinery in the context of TCMTF is explored in
a different document.
ciao
L.
Thanks,
Jose
>-----Mensaje original-----
>De:[email protected]
<mailto:[email protected]>[mailto:[email protected]
<mailto:[email protected]>] En nombre de
>Luigi Iannone
>Enviado el: jueves, 11 de octubre de 2012 9:11
>Para:[email protected] <mailto:[email protected]>
>CC:[email protected] <mailto:[email protected]>
>Asunto: Re: [tcmtf] TCMTF: two kinds of services
>
> Hi Jose,
>
> I would put LISP in two different layers.
>
> You can consider LISP as composed of two main parts:
>
> 1. the encap/decap operations (i.e., the tunnelling)
>
> 2. the mapping system (basically controlling which traffic is tunnelled).
>
> The first can be used at the tunnelling layer, beside GRE/MPLS/L2TP (having
> multiple options is always a good thing ;-) ), to perform the tunnelling. Some
> parts of the LISP header might be exploited to carry TE information (e.g., the
> Instance ID can be used as QoS tag).
>
> The second can be used at Mux/Demux layer. The LISP mapping system
> might be extended to perform tcmtf negotiation by leveraging on a
> particular form of mappings. The advantage is that there is already quite a
bit
> of signalling machinery out there that can be easily re-used/extended.
>
> Ciao
>
> Luigi
>
>
> On 10 Oct. 2012, at 13:48 , Jose Saldana <[email protected]
<mailto:[email protected]>> wrote:
>
> > Luigi,
> >
> > Where would you include LISP here? With GRE/L2TP, or with MPLS?
> >
> > We are also thinking about including other tunneling schemes, as VLAN,
> etc.
> >
> > (to be seen in Courier letters)
> >
> > TCP/IP UDP/IP RTP/UDP/IP
> > \ | /
> > \ | / ------------------------------
> > \ | /
> > Nothing or ROHC or ECRTP or IPHC header compressing layer
> > |
> > | ------------------------------
> > |
> > PPPMUX or other mux protocols multiplexing layer
> > |
> > / \ ------------------------------
> > / \
> > / \
> > GRE or L2TP \ tunneling layer
> > | MPLS
> > | ------------------------------
> > IP
> >
> > Thanks,
> >
> > Jose
> >
> >> -----Mensaje original-----
> >> De:[email protected]
<mailto:[email protected]>[mailto:[email protected]] En nombre
> >> de Luigi Iannone Enviado el: miércoles, 10 de octubre de 2012 10:23
> >> Para:[email protected] <mailto:[email protected]>
> >> CC:[email protected] <mailto:[email protected]>
> >> Asunto: Re: [tcmtf] TCMTF: two kinds of services
> >>
> >> Hi All,
> >>
> >> On 10 Oct. 2012, at 09:58 , Jose Saldana <[email protected]
<mailto:[email protected]>> wrote:
> >>
> >>> - Tunneling will be used to send the multiplexed packets end-to-end.
> >>> The options in this layer are L2TP, GRE and MPLS.
> >>>
> >>
> >> I was wondering if the group might be interested in exploring also
> >> other
> > form
> >> of tunnelling, e.g., LISP.
> >>
> >> Obviously, since LISP is an experimental effort, it cannot be tagged
> >> as
> > "best
> >> current practice" and included in document (A).
> >>
> >> However, the group may have a slightly "larger" scope and include
> >> alternative forms of tunnelling (and/or multiplexing and/or header
> >> compression) (may be documented separately).
> >>
> >> Thoughts?
> >>
> >> ciao
> >>
> >> Luigi
> >>
> >>
> >>
> >> _______________________________________________
> >> tcmtf mailing list
> >>[email protected] <mailto:[email protected]>
> >>https://www.ietf.org/mailman/listinfo/tcmtf
> >
>
> _______________________________________________
> tcmtf mailing list
>[email protected] <mailto:[email protected]>
>https://www.ietf.org/mailman/listinfo/tcmtf
_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp
