1. Pages
2. …
3. Open Transport Configuration and Control Home
4. TAPI
5. TAPI Call-in Details and Notes

2021-09-14 TAPI Meeting Notes

All

• Invited Ramon, Nigel, Ronald, Pedro, Arturo. Please others interested to contact Andrea.

21 Sep 2021 TAPI weeky call

Preliminary agenda:

• Continue the review of TAPI 2.3.1 pre-release (decoupling FM from Notification and Streaming)
• Presentation of  otcc2021.JS.001_tapi-security.docx
• TAPI 2.4 plan: otcc2020.ND.013_Tapi-NotesOn2.4.pptx contribution
• Loopback Use Case
• AccessPort/connectorPin is read-only, check whether the object keys of ConnectorPinAddress are necessary

Arturo indicates that the next version of MUST_Optical_Controller_NBI_Requirements_Document_v10_FINAL_VERSION_WEBSITE.pdf will include a backlog of prioritized use cases.

• The publication is foreseen in one month or so.

Andrea Mazzini shows the main contents of pull request 515.

• Discussion on the OAS generation, the related script is becoming complex / circular because:
  • TapiFm imports TapiStreaming
  • TapiStreaming imports all core TAPI modules (tapi-common, tapi-topology, tapi-connectivity, tapi-equipment, tapi-oam, tapi-path-computation,tapi-virtual-network)
• Agreed to create two new modules, TapiFmNotif and TapiFmStreaming to better decouple TapiFm from both TapiNotification and TapiStreaming.
  • These new modules will include only the fault management augments of notification and stream.
  • Agreed to commit these modifications within pull request 515.

• Discussion on "feature" and "if-feature".
  • Andrea Mazzini informs that in MEF Resource Model Project the usage of these statements has been agreed to define some macro features, e.g. "full topology management".
  •  Karthik Sethuraman we may explore the introduction of "connectivity" and "oam" features. For further analysis.

• Arturo asks whether there are news on yang2oas tool.
  • No news so far.
  • Karthik Sethuraman answers that there is some interest on MEF side.
• A comment has been recently added to TAPI issues 494:
  • "I can see that OAS folder is still present in future r2.3 release for TAPI. Again, if we are publishing something that we know is wrong to our own specifications, sounds like a very bad idea."
  • Agreed to update the release disclaimer as follows:
    • TAPI OpenAPI Specification - TAPI OAS (OpenAPI Specifications) included in this TAPI release (v2.3.1) are an informative part of the TAPI SDK and intended to support TR-547 Use Cases, where a subset of mandatory API is specified.

• Agreed to add in next week agenda the otcc2020.ND.013_Tapi-NotesOn2.4.pptx contribution.
• Discussion on OamJob, agreed that it can be applicable to almost all management items, we may explore a more versatile model.
• Andrea Mazzini informs that in MEF Resource Model Project has been proposed to apply the "if-feature" statement for "RPC support".
  
  • The team confirmed the decision to remove all RPCs from TAPI 2.4 version.

Huy Tran Andrea Mazzini 

 Andrea Mazzini presents otcc2021.AM.001_TAPI-GNPy-extensions.pptx

• Huy Tran points out that new parameters have been defined related to Raman fiber impairments (co-effects) for a more accurate estimation.
  • Explore the model solution, e.g. a new class which augments OTS/OMS links with Raman amplification.
  • The defined data structure is an array of 96 coefficients, each one applicable to one frequency/spectrum value.
• Recalled that all the proposed "GNPy classes" can be considered as profiles, i.e. common for more instances of links, transceivers, amplifiers, etc.