Re: [Pce] Adopting draft-crabbe-pce-stateful-pce-02 as a PCE WG document

[Ramon Casellas <[e-mail]>]

El 02/02/2012 9:14, JP Vasseur escribió:
> Dear all,
>
> This document has been presented twice and certainly still deserves more discussion, but it seems that there is enough interest to poll the Working Group
> for WG adoption. Please let us know by Feb 19th noon ET, whether you are in favor/opposed of adopting draft-crabbe-pce-stateful-pce as a WG document.

Dear PCErs

I share Adrian's comments. The stateless PCE as a functional component 
of a control plane is well understood and I think that, in view of 
potential &  new use cases such as software defined networks/openflow 
and other control & management schemes, there is room to take the PCE a 
step further, not necessarily tied to a GMPLS control plane.

I cannot comment on the concrete, technical details of the draft itself 
yet, as I just read it globally (I plan to read it carefully when time 
permits, most surely before Paris).

To be clear, yes I support the adoption of the draft.

However, as a minor comment, the stateful PCE is briefly covered in 
RFC4655, and in more depth in sections 2-5 of crabbe's draft, but I 
wonder (I am really not sure myself) whether we would need to split it 
and consider first, a more detailed architecture, functional 
requirements/framework, use cases (i.e. no signaling, direct vertical 
configuration of the forwarding state) and architecture, and once agreed 
on those, propose protocol extensions.  Is the stateful PCE a wide 
enough topic to go this path? The draft seems to bundle a full solution 
:) -- I guess this is more an open question rather than a comment on the 
draft itself

Thanks
R.


-- 
Ramon Casellas, Ph.D.
Research Associate - Optical Networking Area -- http://wikiona.cttc.es
CTTC - Centre Tecnològic de Telecomunicacions de Catalunya, PMT Ed B4
Av. Carl Friedrich Gauss, 7 - 08860 Castelldefels (Barcelona) - Spain
Tel.: [phone] -- Fax. [phone]