ietf-ccamp-wg / draft-ietf-ccamp-optical-impairment-topology-yang

8 stars 10 forks source link

Replace OMS/OTS link with OMS/OTS Media Channel Group (see: ITU-T Recommendations G.872 and G.807) #101

Closed dieterbeller closed 2 years ago

dieterbeller commented 2 years ago

ITU-T G.872 (12/2019)

8.4.1 OMS and OTS media channel group The OTS media channel group (MCG) and OMS MCG are described in clause 11 of [ITU-T G.807] as topological constructs that are used for management control purposes. The management and monitoring of these entities is also described in clause 11 of [ITU-T G.807]. NOTE – The terms OMS media link and OTS media link used in the previous version of this Recommendation have been replaced with the terms OMS media channel group (MCG) and OTS MCG in [ ITU-T G.807].

The draft text needs to be aligned with the latest revisions of the referenced ITU-T Recommendations G.807 and G.872.

italobusi commented 2 years ago

I have a mixed feeling about this ...

On one side, I can understand that using the terms OTS/OMS MCG would align the terminology used in the draft with the ITU-T G.807 terminology

On the other side, the OTS/OMS MCG from a topological perspective are modelled as links (at least the OMS MCG) so I am not sure whether the term MCG to represent a link would cause some confusion

dieterbeller commented 2 years ago

Here is the text update I prepared with the objective to align with ITU-T G.807 and describe how the new terms map to the IETF TE-topology terminology:

2.3. OMS Media Channel Group

According to [G.807] and [G.872], an OMS Media Channel Group (MCG) represents a topological construct between two WDM-TE-nodes. In TE- topolgy terms, an OMS MCG corresponds to a WDM TE-link interconnecting two WDM-TE-nodes. Specifically, it originates at the ROADM in the source WDM-TE-node and terminates at the ROADM in the destination ROADM including the Booster Amplifier (BA) and the Pre- Amplifier in the WDM-TE-nodes as well as In-Line Amplifiers (ILAs).

OTS Media Channel Group represents a link:

 (i)  between a WDM-TE-node's BA and the adjacent ILA,
(ii)  between a pair of ILAs,

(iii) between an ILA and the adjacent WDM-TE-node's PA.

An OMS MCG can be decomposed into a sequence of OTS MCGs of type (i), (ii), and (iii) as listed above. An OMS MCG (TE-link) provides the optical impairment data for its elements defined in the layer-0 topology YANG model and can be retrieved by a network controller.

For the sake of optical impairment modeling, an OMS MCG can also be decomposed into a sequence of elements such as BA, fiber section, ILA, concentrated loss, and PA.

An OMS MCG is terminated on both ends by a link termination point (LTP) as defined in [RFC8345]. Links in optical transport networks are typically bidirectional but have to be modeled as a pair of two unidirectional links following the [RFC8345] modeling approach. Unlike TE-links, which are unidirectional, the LTPs on either end of the TE-link pair forming the bidirectional link, are bidirectional as described in [I-D.ietf-teas-te-topo-and-tunnel-modeling] and the pair of unidirectional links are connected to the same bidirectional LTP on either end of the link pair.

ggrammel commented 2 years ago

few nits: The OMS-MCG starts from the ROADM element in the WDM-TE-node, but the description of OTS-MCG doesn't cover the link between ROADM element and Booster/PA. To cover this, a case iv should be added. (see G.807 Fig 11-1)
(iv) between WDM-TE-nodes ROADM and the BA/PA

last paragraph: [RFC8345] does not specifically define LTPs. It defines Termination Points (TPs) in a generic manner. So "LTP" would better be replaced by "TP"

sergiobelotti commented 2 years ago

few nits: The OMS-MCG starts from the ROADM element in the WDM-TE-node, but the description of OTS-MCG doesn't cover the link between ROADM element and Booster/PA. To cover this, a case iv should be added. (see G.807 Fig 11-1) (iv) between WDM-TE-nodes ROADM and the BA/PA

last paragraph: [RFC8345] does not specifically define LTPs. It defines Termination Points (TPs) in a generic manner. So "LTP" would better be replaced by "TP"

@ggrammel @dieterbeller : thanks Gert to pointed out, we need to substitute RFC8345 with RFC8795, since LTP it is a a te-topology defined entity.

dieterbeller commented 2 years ago

Fixed by PR #105