| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
Alper,
it really sounded like ES and CS @ L2 are “the” solution so there is no need for ES and CS at L3, that’s why I reacted. If we agree on the need for ES and CS @ L3, then I am fine, and I would avoid further references to what is being done or not @ L2, since we’re here to work on the L3 solution. Whether the same protocol needs to provide all three service or not is another question, and I would not discuss its merit at this point (that’s more of DT decision than a chair suggestion).
Stefano
From: Alper
Yegin [mailto:alper.yegin@yegin.org]
Sent: Wednesday, March 21, 2007 6:58
AM
To: Stefano Faccin;
'Telemaco Melia'
Cc:
STDS-802-21@listserv.ieee.org; mipshop@ietf.org; 'Vijay Devarapalli'
Subject: RE: [Mipshop]
DHCP
Stefano,
I’m not denying the need for a L3 ES and CS transport.
I’m saying L2 ES and CS transport is also one of the accepted scenarios.
Are we not on the same page?
Alper
From: Stefano
Faccin [mailto:smfaccin@marvell.com]
Sent: Wednesday, March 21, 2007 2:15
PM
To: Telemaco Melia; Alper
Yegin
Cc:
STDS-802-21@listserv.ieee.org; mipshop@ietf.org; Vijay Devarapalli
Subject: RE: [Mipshop]
DHCP
Indeed. However, I need to add something. Alper, can you help me understand in which forum or framework or circle it is an accepted scenario that ES and CS are taken care at L2? That may be true for some (few) access technologies, but cannot be generalized, therefore there is the need for a L3 solution. As a matter of fact, there is plenty of discussion going on on why it is useful to have it @ L3 for many technologies, in particular to have a technology independent solution for multi-mode terminals and networks.
If we want to talk about L2 solution, there are technologies (e.g. 802.11) that have solutions @ L2.
Stefano
From: Telemaco
Melia [mailto:telemaco.melia@netlab.nec.de]
Sent: Wed 3/21/2007 6:10 AM
To: Alper Yegin
Cc: STDS-802-21@listserv.ieee.org;
mipshop@ietf.org; 'Vijay Devarapalli'
Subject: Re: [Mipshop]
DHCP
Hi
Alper,
I would suggest to wait for tomorrow's DT
presentation.
telemaco
Alper Yegin wrote:
>
There have been already proposals using DHCP for
802.21 services
>
discovery.
>
> The DT is currently
considering different scenarios and associated
>
requirements targeting a single solution.
>
This might change if complexity increases too
much.
>
>
> I don't understand why designing one protocol that
does three things (IS,
> ES, CS) is better than using an existing protocol
(DHCP) for what it has
> used for (configuration [IS]), and designing
another protocol that takes
> care of the other two (ES,
CS).
>
> I'm not aware of any technical reason that requires all
three
> functionalities to be bundled up in single protocol. And in fact,
I see
> reason to separate them. It is an accepted scenario that ES and CS
are taken
> care of by the L2 in some networks, leaving IS to
L3.
>
> Comments?
>
>
Alper
>
>
>
>
--
Telemaco
Melia
telemaco.melia@netlab.nec.de
Research Staff Member
Tel: +49 (0) 6221 4342-
142
Network Laboratories
Fax: +49 (0) 6221 4342- 155
NEC
Europe Ltd.
Web: http://www.netlab.nec.de
Network
Laboratories
Kurfuersten-Anlage
36
69115
Heidelberg
GERMANY
NEC Europe Limited | Registered
Office: NEC House, 1 Victoria Road , London W3 6BL | Registered in England
2832014
_______________________________________________
Mipshop
mailing list
Mipshop@ietf.org
https://www1.ietf.org/mailman/listinfo/mipshop