[Seamoby] RE: what is edge signalling? (was: RE: [NSIS] Re: Comments on dra ft-westphal-nsis-qos-mobileip-00.txt)

["Hancock, Robert" <[e-mail]>]

hi gary,
 
my comments are purely on draft-westphal-nsis-qos-mobileip and associated email discussions (since that's been the only input I have seen recently which explicitly discusses the relationship between CT and NSIS, and which I had worries about). 
in particular, any comments I make about MN-AR signalling are really to emphasise the difference between the cases of
a) not having CT signalling that carries QoS information MN-AR (if there is or isn't other CT signalling I don't really care), and
b) having CT signalling that does carry QoS information MN-AR, which seems to be the assumption of the mentioned draft.
 
our recent framework activities assumed (a). if you're saying that the seamoby w.g. requirements also impose (a), that's fine by me. i also don't want to re-open old discussions (certainly not on the nsis list, anyway).
 
i suspect we're basically in agreement about NSIS/CT relationships: after a handover, if CT happened, it should be possible to have less NSIS signalling on the on-path segment which goes over the air.
 
cheers,
 
robert h.
 
PS referring to a previous email, i think the view that NSIS is "likely to be of limited use in mobility situations" is rather pessimistic. i would rather say that if you want the resources signalled for by NSIS to be available seamlessly over handovers, you'll need something else as well, which could be CT. the basic point seems to be that NSIS shouldn't try to solve the cross-path context/state transfer problem - any more than CT should try to solve the along-path context/state establishment problem. 

-----Original Message-----
From: Gary Kenward [mailto:[e-mail]]
Sent: 05 July 2002 15:35
To: Hancock, Robert; '[e-mail]'; [e-mail]; [e-mail]; [e-mail]
Cc: [e-mail]; [e-mail]; '[e-mail]'
Subject: RE: what is edge signalling? (was: RE: [NSIS] Re: Comments on dra ft-westphal-nsis-qos-mobileip-00.txt)



Robert: 

  In your discussion of CT, you are presupposing a wg solution 
for CT that does not exist. The work on the CT protocol has not begun 
in Seamoby. Indeed, if you examine the CT requirements, 
signalling from the MN to the access network is prohibited. The 
reason being is that the sole purpose of CT is to enable a fast, smooth, 
seamless handover. Signalling over the air simply takes too much time, 
and the value of signalling from MN to AR has never been clearly identified. 

  This issue was debated extensively in the CT design group and on 
the Seamoby list. I understand that there are people who feel otherwise, 
and please, let's not regurgitate the arguments here. As the requirements 
currently stand, CT is an AR to AR protocol. 

  One source of confusion might be the work on the discovery protocol. 
It is my, admittedly minority opinion, that the use of MN to AR signalling 
to perform CT was moved into the discovery solution. And again, nothing 
can be done between the MN and an AR that cannot be done more efficiently, 
and faster, between two ARs. 

  The way I see the directions of the two solutions lining up so far, 
NSIS and CT have little interaction. NSIS could be invoked if a CT attempt 
fails, but the restriction to have NSIS along the path implies that some 
path must exist before NSIS can be invoked. Thus, NSIS can be used only 
*after* a handover has completed, and thus there is no opportunity for 
interaction. Of course, all of this depends on what you define as a "path" 
- I am assuming that it is the IP forwarding path for the MNs bearer traffic. 


gary 

> -----Original Message----- 
> From: Hancock, Robert [ mailto:[e-mail] <mailto:[e-mail]> ] 
> Sent: July 5, 2002 05:35 
> To: '[e-mail]'; [e-mail]; 
> [e-mail]; [e-mail] 
> Cc: [e-mail]; [e-mail] 
> Subject: RE: what is edge signalling? (was: RE: [NSIS] Re: Comments on 
> dra ft-westphal-nsis-qos-mobileip-00.txt) 
> 
> 
> Hi all, 
> 
> I don't think there's any disagreement that CT is a tool that 
> NSIS could use (as an optimisation). 
> 
> What's entirely unclear to me is what sort of tool CT is: is 
> it a hammer or a screwdriver or a combination of the two? I 
> guess this is more a question for the seamoby people, but we 
> need to know the answer to know how to use it. 
> 
> More formally, there seem to be two extremes: 
> 
> a) NSIS is used for MN-AR resource (QoS) signalling, and 
> AR-rest of world signalling; CT can be used to transfer state 
> (both resource state and NSIS protocol state) between ARs at 
> handover, but if this involves any MN-AR messages (e.g. CTIN) 
> they aren't QoS related. NSIS-CT interaction can occur at the 
> ARs to do state transfer at handover time. 
> 
> b) CT messages (e.g. CIN) can be used by the MN to set up QoS 
> context in the AR, which it can then later transfer to a new 
> AR. NSIS is used to signal QoS for AR-rest of network. 
> NSIS-CT interaction takes place at the AR to interwork flow 
> setup and resource descriptions, and this interaction takes 
> place at flow setup time. (Essentially, CT provides a QoS 
> signalling protocol for the edge of the network, which NSIS 
> would have to interwork with, in the same way that it has to 
> interwork with RSVP/Intserv.) 
> 
> our framework draft-hancock-nsis-fw-00.txt basically assumes (a). 
> this draft-westphal-nsis-qos-mobileip-00.txt basically 
> assumes (b) so far as I can tell. 
> 
> I'm not trying to argue which of (a) or (b) is 'correct' at 
> this stage, just establish that they are not the same. (And 
> so we probably have to make a choice, although I suppose you 
> could allow both models...) 
> 
> Cheers, 
> 
> Robert H. 
> 
> > -----Original Message----- 
> > From: Marcus Brunner [ mailto:[e-mail] <mailto:[e-mail]> ] 
> > Sent: 05 July 2002 10:05 
> > To: [e-mail]; Hancock, Robert; [e-mail]; 
> > [e-mail] 
> > Cc: [e-mail]; [e-mail] 
> > Subject: RE: what is edge signalling? (was: RE: [NSIS] Re: 
> Comments on 
> > draft-westphal-nsis-qos-mobileip-00.txt) 
> > 
> > 
> > I see CT as an optimization only and not required for NSIS. 
> > However, since 
> > there is a tendency to support only on-path signaling, this 
> > would not work 
> > anyway (because CT would need off-path signaling for at least 
> > some time). 
> > 
> > Marcus 
> > 
> > --On Donnerstag, 4. Juli 2002 11:33 +0300 
> > [e-mail] wrote: 
> > 
> > > Hi Robert, 
> > > 
> > >> On the 'edge signalling' case, I guess the main point is that 
> > >> I'm confused about the scope of seamoby CT. Specifically 
> > >> regarding MN-AR signalling (no issue about the AR-AR part): 
> > > 
> > > From my viewpoint, CT is not required for NSIS. I view CT as 
> > > one tool that an NSIS solution could use.  So, I definately think 
> > > that at some point, we will need some sort of applicability 
> > > statement describing how to use CT for NSIS. 
> > > 
> > > John 
> > > 
> > > _______________________________________________ 
> > > nsis mailing list 
> > > [e-mail] 
> > > https://www1.ietf.org/mailman/listinfo/nsis <https://www1.ietf.org/mailman/listinfo/nsis>  
> > 
> > 
> > 
> > -------------------------------------- 
> > Dr. Marcus Brunner 
> > Network Laboratories 
> > NEC Europe Ltd. 
> > 
> > E-Mail: [e-mail] 
> > WWW:    http://www.ccrle.nec.de/ <http://www.ccrle.nec.de/>  
> > personal home page: http://www.brubers.org/marcus <http://www.brubers.org/marcus>  
> > 
> > 
============================================================== 
========================== 
Permission is hereby granted to pass the contents of this 
communication to third parties and any restrictions regarding 
confidentiality do not apply.