Skip to end of metadata
Go to start of metadata




  • Admin
  • Optical Impairments review
  • Layer Protocol Name and Qualifier of CSEP
  • DSR UNI Model

Agreed Items & Priority

Discussion items

10 minsAdministrative



  • Scheduled two additional, official, dedicated TAPI calls:
    • Wed 13:00-15:00 and Thu10:00-12:00 CEsT

Asked to ONF to extend the additional, official, dedicated TAPI calls till July 2022.

This week the next calls are on Wed 12-15; Thu 10-12; Fri 11-13 CEsT

Next week: Mon 16, 14-16; Wed 18, 12-15 CEsT

  • Andrea Mazzini will send the invitation to anyone who wants to participate.

TAPI weeky call canceled

TAPI weeky call

Preliminary agenda:

  • Continue the Optical Impairments review
  • Check the details of DSR UNI model
80 mins

Optical Impairments review


Andrea Mazzini shows the updated diagrams with ROADM Path impairments.

  • Andrea Mazzini are ROADM Path Impairments to be considered unidirectional?
  • Esther explains that the ROADM Path Impairments augment the connectivity-matrix defined in RFC 8795, hence the impairment directionality is inherited:

  • Esther adds the consideration that the internal ROADM structure could be very complex, while the connectivity-matrix provides the appropriate abstraction for the purpose of impairments evaluation.
    • Also clarified that the ROAM Path Impairments do not include the Amplifier impairments, they are outside the box of the ROADM connectivity-matrix.
  • Nigel Davis the semantic is analogous to the TAPI Node Rule Group, the switching layer being Media Channel:

  • Andrea Mazzini hence the Node Rule Group shall be defined between MC NEPs, which is in line with the exclusion of amplification functions, modeled at OMS layer CEP.
    • The Node Rule Group shall refer to the ROADM Path Impairments profile. It is a specific realization of the Transfer Cost.
    • The same Node Rule Group instance may refer to more ROADM Path Impairments profiles in case of different impairments on different bandwidths.
    • Agreed to add an optical impairments forwarding rule, to allow for a finer granularity of description, e.g.:

NodeRuleGroup 1
MC NEP of degree 1
MC NEP of degree 2
MC NEP of degree 3
MC NEP of degree 4

NodeRuleGroup 2
MC NEP of degree 1
MC NEP of degree 2
pointing to the ROADM profile1

NodeRuleGroup 3
MC NEP of degree 1
MC NEP of degree 3
pointing to the ROADM profile2


Andrea Mazzini shows the Fiber + Concentrated Loss diagram:

  • Agreed that the right place where to add these impairments is the OTS_MEDIA CEP.
    • Considering the OSC, another dedicated CEP could be made available on the same NEP, hence better the OTS_MEDIA CEP.
  • Agreed that the Fiber Profile shall be referred by the AbstractStrand.
30 mins



Andrea Mazzini presents some slides on the DSR UNI model. The idea is to keep a single NEP-SIP couple for a port which can support either DSR or ODUk rates.

  • Agreed that instead of DSR it is better PHOTONIC_MEDIA (and in future maybe also ELECTRIC_MEDIA).
  • Discussion on the "simplified" model, agreed a further simplification, from the expanded model:

  • To the compact model, with only one CEP instance, including 10GE and ODU2 functions: