Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Migrated to Confluence 5.3

Date

17 October 2017

Attendees

Apologies:

Goals

  • Administrative
    • Record this meeting
    • ONF Interim Meeting
    • ITU-T London Meeting
  • Ethernet model for TAPI & MEF (Andrea)
  • Alarm & TCA notification model for the next stage (probable cause, perceived severity etc) (Nigel)
  • AOB

Discussion Items

TimeItemWhoNotes
 Admin- InterimNigel
  • Agenda preparation underway based on Kam's framework
  • 50:50 allocaction between Core and TAPI (OT-IM work is assumed absorbed into TAPI)
  • Lyndon and Nigel are preparing material for CORD Build
    • We may need help with TAPI demo prep and slides as Karthik is out
 Admin - LondonNigel
  • The meeting will be at Ciena London
  • Logistics details will be posted to the ITU-T site today or tomorrow
 Ethernet modelAndrea
  • Andrea summarised the model

  • Open points:
    • Adaptation of G.8052 constructs to the TAPI structure and shared a view of the Papyrus model
      • EthernetAdaptationPac v EthernetCtpPac
      • Nigel noted that we need to show the adaptation in detail to understand the problem (as there is an adapter at the top and the bottom of the LP.
      • Italo noted that we need to take care to not have too much detailed structure but agreed with Nigel that we need to understand the detail before we can abstract it
    • Andrea showed a figure "NRM: mapping on SIP and SEP (UNI-N/EVC)" which showed substructuring of the LTP procressing
      • Nigel noted that a single LP is allowed to hold all aspects of Termination, adaptation, pooling and connectability
      • Nigel noted that an LP can have any subset and LPs can stack hence the figure was valid
    • Andrea noted the NodeEdgepoint and the ConnectionEndPoint partitioning of functionality needs to be considered carefully
      • NEP is partly termination and partly topological pooling... the balancing is difficult
      • Italo noted that the media should not be visible at the ethernet connection layer
      • Nigel asked whether the challenge was the degree of collapsing, e.g. where there is a stack of dull layers from physical to the first layer of interest.. all these properties could be in an NEP
    • Bernd/Italo/Nigel questioned the LAG aspect of the figure
      • Italo offered to update the figure
      • Bernd noted that there was a level of NEP/CEP missing
      • Italo suggested that OAM needed to be added
      • Bernd asked what frame was being used to monitor
      • Italo explained that this is an untagged frame
    • Bernd showed a view of LAG
      • Andrea suggested that this is not his scope
    • Nigel summarized that the issue was that the diagram appeared to be making a statement about the relationship between the C-Vlan layer and the Phy... that was not the intention and hence the diagram either needed to be modified to show a cloud between the two parts, modified removing the phy or be converted into a set of examples with and without LAG.
    • Nigel requested that:
      • Andrea submits the current diagram and model via the minutes DONE
      • Andrea make appropriate mods
      • Italo take Andreas diagrams and assess for further mods
      • Bernd attaches his figure to the minutes DONE
 Alarms and TCA 
  • Not discussed due to time restrictions

Action Items

  •  Italo Busi Take Andreas diagrams and assess for modifications and improvements
  •  Andrea Mazzini Refine figures as per discussion