Re: [CCAMP] Webex meeting invitation: Optical Impairments - occurrence of week 40

["Beller, Dieter (Nokia - DE/Stuttgart)" <[e-mail]>]

Hi Gert, all,

please find my comments inline below.

Thanks,
Dieter

On 04.10.2019 19:43, Gert Grammel wrote:
Sergio, Italo,

Thank you for the summary. Two clarifications:

  *   In the report it is stated that the LLCL is a Link between OT and the line (crossing ADD-DROP block) and is a candidate to be used to attach the impairment parameters on this. However further down it is said a domain controller would not expose it (which basically creates an aggregated model). Since the preference was for a disaggregated model, there has been no final conclusion on what exactly to expose and what not.

[DB>] As far as I remember, we agreed that the SDN controller should abstract, if possible, a dis-aggregated ROADM in the same way as it represents an integrated ROADM.
[DB>] This means that the optical impairments imposed by any physical component between the optical transponder's line port and the OMS-link should be modeled by
[DB>] augmenting the LLCL (see: https://tools.ietf.org/html/draft-ietf-teas-yang-te-topo-22#section-3.8).


  *   The group also discussed that add/drop would only need to be exposed in case of a domain Boundary. My intervention aimed to clarify that this there are other use cases beyond domain boundaries that benefit from exposing such link between Add/Drop and OT. Given the preference expressed by the group for a disaggregated model, it seems pre-mature to state “In the first phase, we will consider the case where it is not exposed.”

[DB>] The fiber link between the OT and the Add/Drop typically is a (single-channel) fiber link. In many cases it is so short that the optical impairments of this passive fiber
[DB>] are negligible. As explained above, we are going to investigate whether the optical impairments imposed by the physical components between the optical transponder's
[DB>]  line port and the OMS-link can be abstracted into a set of impairment attributes (LLCL augmentation) such that it is not necessary to model each physical component
[DB>] separately. Only if there is a domain boundary between the optical transponder and the Add/Drop block, it is probably necessary to expose that single channel fiber
[DB>] link.


Gert



From: "Belotti, Sergio (Nokia - IT/Vimercate)" <[e-mail]><mailto:[e-mail]>
Date: Friday, October 4, 2019 at 18:16
To: Gabriele Galimberti <[e-mail]><mailto:[e-mail]>, Daniele Ceccarelli <daniele.ceccarelli=[e-mail]><mailto:daniele.ceccarelli=[e-mail]>, CCAMP <[e-mail]><mailto:[e-mail]>
Cc: LE ROUZIC Esther OLNC/OLN <[e-mail]><mailto:[e-mail]>, Julien Meuric <[e-mail]><mailto:[e-mail]>, Italo Busi <[e-mail]><mailto:[e-mail]>, Gert Grammel <[e-mail]><mailto:[e-mail]>, Dieter Beller <[e-mail]><mailto:[e-mail]>, "Griseri, Enrico (Nokia - IT/Vimercate)" <[e-mail]><mailto:[e-mail]>
Subject: RE: [CCAMP] Webex meeting invitation: Optical Impairments - occurrence of week 40

Hi all,

Related to the meeting on Wednesday October 2nd ,
Italo and myself put together some notes (attached), in the view to have common understanding of the topics been touched and so progressing of discussion for next meeting.

Please look at the attached word and provide your comments if any to complete the summary.

Thanks
Italo and Sergio


Sergio Belotti
Senior System Engineer and Standardization Architect
IP/Optical Networks, Optics BU
Nokia
M: [phone]
Via Energy Park, 20871 Vimercate (MB) , Italy
[e-mail]<mailto:[e-mail]>