Re: How can I get Waveband Label?

[Dimitri.Papadimitriou@alcatel.be]

hi, we can consider several approaches (at least i see three of 
them), the current definition of waveband switching (see [gmpls-
arch] and [gmpls-sig]) refers explicitly to inverse multiplexing 
mechanism and wavelength concatenation (“contiguous” lambdas 
switching) were one can request a 10 Gbps lsp while the underlying 
(physical) wavelength operates at 2.5 Gbps. 

this definition does not explicitly considers the (3rd) approach 
where band has a physical signification, i.e. where the interface 
would be waveband switch capable (wbsc) through a waveband lsp
encoding, to respond to your question; this has been the rationale
for the following i-d:
http://search.ietf.org/internet-drafts/draft-douville-ccamp-gmpls-waveband-extensions-01.txt

where this third approach has been developed but doesn't seem to 
have received yet a tremendous amount of interest from the ccamp 
wg community, and far from me the idea to say that the method
proposed in this i-d is the only & unique solution

however, it is also true that one could find some artifacts in order
to make a lambda request becoming appearing as a wb request for 
instance by using resource coloring, but in this case the "waveband 
label" may become questionable, nevertheless even if solving one 
part of the problem such method become very complex (see also the 
thread on resource classes initiated by sven) to use in multi-
region environment since only one lsc region has been defined (we 
don't have for the moment as it is the case for psc, four values), 
by considering additional levels (lsc-1, lsc-2, etc.) one may more
than certainly relax this complexity

i would be probably more than useful to have other opinions on 
this, any other views ? at least one should open the debate since
multi-switching lsr's receive more and more attention these days

thanks,
- dimitri.




Wataru Imajuku wrote:
> 
> [ post by non-subscriber.  with the massive amount of spam, it is easy to
>   miss and therefore delete mis-posts.  so fix subscription addresses! ]
> 
> Hi, all.
> 
>   I cannot find proper solution to realize waveband label request.
>   It seems that latest version of generalized label request object with the
> LSP Enc. Type and Switching Type can not request Waveband Label from
> downstream node.
> 
>   Because, neither of LSP Enc. Type and Switching Type has identifier to
> request waveband label to downstream node. Especially, Switching Type
> defines Lambda Switch Capable (LSC), but does not define Wavenband Switch
> Capable.
> 
>   How can I request Waveband Label?
> 
> W. Imajuku
> 
> ---------------------------------------------
> Wataru Imajuku
> Research Engineer@NTT Network Innovation Labs.
> 807A, 1-1 Hikari-no-oka, Yokosuka, Kanagawa, Japan
> Zip: 239-0847
> E-mail: imajyuku@exa.onlab.ntt.co.jp
> TEL: +81-468-59-4315
> FAX: +81-468-59-3396

-- 
Papadimitriou Dimitri 
E-mail : dimitri.papadimitriou@alcatel.be 
Website: http://www.rc.bel.alcatel.be/~papadimd/index.html
Address: Alcatel - Optical NA, Fr. Wellesplein, 1 
         B-2018 Antwerpen, Belgium
Phone:   Work: +32 3 2408491 - Home: +32 2 3434361