Re: I-D ACTION:draft-andersson-mpls-g-chng-proc-00.txt

[Gert Grammel <Gert.Grammel@alcatel.de>]

Neil,

I think you haven't got my point about call/connection separation. Please have
also a look to my respone to Eve. There can be a discussion about
call/connection separation once there is a kind of shared view on 'why it is
needed'. In
http://www.ietf.org/internet-drafts/draft-ietf-ipo-carrier-requirements-05.txt
the only 'motivating sentence I could find was:

     To support many enhanced optical services, such as scheduled bandwidth
     on demand, diverse circuit provisioning and bundled connections, a
     call model based on the separation of call control and connection
     control is essential.

You gave some practical examples on where you see the need for that. Why not
putting it into a document and share it with everybody?

I also want to thank you to share with us your choice to go for a L0/1
functionality. This is a very clear statement.


Regards

Gert

neil.2.harrison@bt.com wrote:

> Hi Gert, I think some clarifications are in order here to make the BT
> position (ie not just the NH position) clear.  Please see below.  regards,
> Neil
>
> Gert Grammel wrote 11 March 2003 15:47 to Monica Lazer:
>
> <snipped>
> > Look at the hottest issue under discussion: We are talking a
> > lot about UNI
> > features like call/connection separation. Neither Mark nor
> > Neil see the UNI
> > becoming service relevant at any time soon - so why is it
> > under discussion right
> > now? Is this your priority 1 issue in ITU-T? Is Mark or Neil
> > in a minority
> > position in ITU-T? Referring to the Yokohama Meeting this was
> > basically
> > Kireeti's request: Please explain what you need and for what purpose.
>
> NH=> A UNI at L1/0 is *not* something we see as important anytime soon.  We
> also see no need for coupling with L2/3 (either commercially or
> technically).  Thus we want to be able to choose best-of-breed functionality
> for L1/0.  We *do* want call/connection separation however, and note this is
> independent of the UNI...here is an extract from a colleague of mine on the
> ITU lists which sort-of explains why:
> "Anyone care to disagree with the notion of a call handle for SPCs in the
> ITU-T as a means of bundling under one service instance a customer record
> that handles
> multiple connections e.g. Virtual Concatenation, with or without LCAS as a
> valid application?
> Anyone care to disagree that we might want to retrieve information on an SPC
> using a single identifier (the call) that shows all connections associated
> with that call and the performance of each connection and/or start end of
> individual connections?
> Anyone care to diagree that the call/connection model cannot be used for
> restoration of SPCs?
> As for those that want to do UNI on you go...but its not on BT priority
> list."
> >
> <snipped to end>

--
Alcatel Optical Network Division    Gert Grammel
Network Strategy                    phone: +49 711 821 47368
Lorenzstrasse 10                    fax: +49 711 821 43169
D-70435 Stuttgart                   mailto:Gert.Grammel@alcatel.de