Child pages
  • 2018-03-29 OIMT Meeting Notes
Skip to end of metadata
Go to start of metadata

Date

29 March 2018

Attendees

Apologies

Nigel Davis

Goals

Discussion Items

TimeItemWhoNotes
IM-DAdminKam
  • Status of upcoming F2F meetings
    • The confirmation of the December Australia meeting is pending on whether there will be a CORD Build at the year-end of 2018.
    • Action Item - Kam: Ping Cassandra to find out if there is going to be a CORD Build meeting this year.
  Kam
  Andrea
  • MEF Liaison Statement
    • LS from MEF on Subscriber L1CS Attributes draft
      • Commented and clarified on the term "Service"
      • The service attributes are supposed to be in the attachment of the LS, but there is no attachment.
        • Malcolm noted that the MEF specification was attached to the liaison also to ITU-T that was posted as TD184/3 in SG15

      • Action item - Andrea Mazzini: To give an overview of the MEF Subscriber L1CS Attributes Draft in the next week call.
   
  • Topics for next week OIMT discussion
    • Lifecycle use cases (Malcolm)
    • (skip control model example)
    • Generic UML model header information (Bernd to present)
    • Outgoing LS that were planned at London meeting.
 Packet examplesXiang
  • Comments from 3/27 OT-IM review of oimt2018.XY.001.07_Draft-TR-512.A.6…
    • Figure 4-2, the lower black dot should be white.
    • Figure 4-2 and Figure 4-4: Add the termination triangle to the S-Tag (per the conclusion below regarding S-Tag should be treated as a layer).
    • Action item - Xiang YUN: Update Figure 4-2
      • DONE: use white dot for the lower black dot in Figure 4-2
      • DONE: update Figure 4-2 and Figure 4-4 with termination triangle to S-Tag
    • Comment on Figure 4-1, make 0x88e7 and the type cells consistent on notation, prefer "type"
      • DONE: update 0x88e7 with EtherType
 ControlConstructChris / Malcolm
  • Discussed 3/22 raised comments
    • Concern with the Master & slave names
      • Possible alternative names to consider
        • leader / follower
          leader / minion
          parent / child
    • Need to consider mapping between ConstraintDomain with roles (for a particular role with a particular client/server). This will help explain how to use the model.
      • For v1.4, the basic concept of the model is fine

        • ControlConstruct, ControlConstraint, and the Port of ControlConstraint
      • Consider to have examples of usage of the model in v1.5 A-series, such as representing the Controller per the ITU-T architecture, e.g., Call Controller, Connection Controller,

        • Generic and robust to both SDN control and ASON control
      • The name of abstract mapping entity, ProcessingConstruct or ControlConstruct, seems is ProcessingConstruct
 LifecycleMalcolm
  • Status Updates
    • State transition and condition
      • Currently there is no way for the (controllers of the) supporting entity and supported entity to communicate for state transition gracefully
        • For example, a supporting entity is in PendingRemoval, how can (the controller of) the supported entity communicate to (the controller of) the supporting entity that the supporting entity can be removed.
      • Action item - Malcolm Betts: Work on proposals for discussion next week. (Also update the above problem statement as necessary)

Action Items

  •  

2 Comments

  1. not sure what you mean by "Generic and rubus"

    "Rubus is a large and diverse genus of flowering plants in the rose family, "

  2. I think it was a typo. I think I heard "robust" but miss-typed as Shouldrubus. Thanks for checking the minute.