ONAP R6 Resource IM Call 2019-12-16

General Information:

• Date and Time: 2019, Dec 16th, 10pm~11pm Beijing Time, 9am~10am US Eastern
• Meeting Room: https://zoom.us/j/645982535
• Meeting Recording:  see bottom
• Meeting Chat Log:  see bottom
  

Attendees

• Kevin Scaggs
• Xu Yang
• Jacqueline Beaulac
• Mehmet Toy
• Michela Bevilacqua
• Zu Qiang (Ericsson)
• Benjamin Cheung
• Chuyi Guo

Agenda:

• Agenda bashing - 5 minutes
  • items for checking:
    
    • ves spec
    • PNF instance model
    • modeling glossary update
    • plan for X'mas and new year's eve
• allotted resource (final comments)  - 20 minutes
• Modeling Process discussion - 25 minutes

Material:

• Proposed ONAP Release Process Updates for Information and Data Modeling
• Modeling sub-committee Contributions
• Allotted Resource Revised
• New PNF example

Minutes:

• Agenda bashing:
  • VES specification -
    
    • Kevin has checked with Alok, VES 7.1.1 is expected to be closed one week ago (and Alok is OOF since then)
    • the only impact to IM is to change one attribute which is related to one relationship
  • PNF instance model - 
    • would discuss after the new year
  • Modeling glossary update -
    • DOC TEAM DISCUSSION - Andy & Kevin joined the doc. team discussion last week, spend time showing what had been developed. Well received. Highlighted idea of experimental vs preliminary. Some exp. don't necessarily have def.; incl. exp if it has a definition. exper. is ok to have. This may be just one aspect of the glossary. Acronym is not in an information model.
    • NEXT STEPS - As Next step - will add Stereotype back to the GenDoc script. There will be a follow-up discussion in 2020.
    • Q&A -
      • Q: Is the glossary going to be labeled "Modeling Glossary"? A: probably a more general glossary.
      • Q: The terminology should be anchored across different projects e.g. service might be a different?
      • Q: is a cross-project discussion planned? A: Don't know.
      • Q: Who will OWN the Glossary? A: probably the Docs team w/ the Modeling team being a contributor.
      • Q: e.g. slicing NSMF is basic concept for slicing part; will this be also one of the modeling glossary entries? i.e. Domain-specific concepts should they be included in the Glossary? A: probably not. Chuyi Guo should be analyze some of the specific terms? A: Yes. If an information model is about concepts w/ attributes; should a slice be one of those concepts? If there are other sources of info going into the glossary, even if it is not a class, a S/C could submit the concept to the docs team.  
      • Q: Has doc team requested u/c to provide input for glossary? A: No, but 5G u/c team is planning to have documents for the use case.
      • ACTION: Benjamin Cheung will contact Sofia Wallin to see if U/C teams should contribute to the glossary.
  • Holiday Season plan -
    • most people would be off for holiday in the next two weeks, and Jan 6th is a public holiday in Sweden
    • expect to have three weeks off and potentially use Jan 8th (service IM call time slot) for the next call
• Discussed items:
  • allotted resource 
    • comments on the wiki page:
      • 1) suggest to provide U/C. Kevin would ask Gil if he could provide some.
      • 2) why ANFD and ANFI are lacking attributes? A: would investigate the possible attributes
      • 3) need to differentiate NFInstance and NFDesc classes as their definitions are quite the same now. A: would fix
      • 4) ANF Instance/ANFDRunsOnSharableNFDesc/ANFInstanceRunsOnSharableNFInstance description are suggested to be added. A: would fix
      • 5) ANFDRunsOnSharableNFDesc/ANFInstanceRunsOnSharableNFInstance relationship needs more clarification
      • 6) Q: recursive relationship is expected to be avoided? A: we do have the agreement w.r.t service model, allotted resource is not the same case. Q: how the descriptors are expected to use together? The description is stated as "Anfd is provided by some service instance", how to understand? A: would do refinement, and discuss later
      • additional comments are not reviewed due to time limit
  • Modeling Process discussion
    • not discussed due to time limit

RECORDING