Child pages
  • 2020 OIMT Virtual Face to Face - Week of April 13
Skip to end of metadata
Go to start of metadata

ZOOM URL:  https://www.zoom.us/j/956285179     Meeting ID: 956 285 179

Schedule

In general we need to focus on agreement and actions :

  1. close off 1.5 release
  2. finalize and prioritize 1.5+ release
  3. firm up 2.0 release


Time-slot

4/13 Day 1

GMT Mon

4/15 Day 2

GMT Wed

4/16 Day 3

GMT Thu

4/17 Day 4

GMT Fri

Slot 1

6:00 - 6:50 EDT

Admin, welcome, 2020 Work Plan
  • Use of Job Task to improve Ops Patterns
  • Simplified Spec vs Generalized Spec
Ethernet Examples
  • Constraint Domain Split
  • Model structure
  • Summary
  • Work plan

Slot 2

7:00 - 7:50 EDT

  • Streaming
  • View / Context TAPI implications
Discuss approaches for EquipmentSecurity 

Slot 3

8:00 - 8.50 EDT

  • View / Context TAPI implications
  • Model Structure Update
  • Intent

Topics

Assigned to Slot

(Day.slot)

a or b for 1/2 slot

Topic TitleLeadLength requested
(full or 1/2 slot)
Topic DescriptionMeeting Notes
with Document Links
and action items
1.1
  • Admin, welcome, 2020 Work Plan
KL60 minsBubble chart slide showing work items and their relationship / planned order
  •  Hing-Kam Lam Discuss work plan with TAPI, in particular for T35 (LTP Port) and T41 (Identity model)

Master list/plan: Planning & Work Items 


1.2.a
  • Streaming
ND30 mins

Note: Moved from 2.1a as this feeds into View Context (ND)

oimt2020.ND.005-Streaming.pptx

Agreed to use the spec approach. 

Not to specialize ControlConstruct for streaming. 

  • Nigel Davis Streaming: Not to use the raw term "Log", stream log, filter log, 
  • Nigel Davis Streaming: Clarify multiplicity between ControlPort and StreamHandler in the context of multiple clients
  • Nigel Davis Streaming: Once change to the Spec approach, check what the impact on TAPI is. Note that the spec approach of streaming is about "How", not "What".

Agreed that in this release, we assume reliable streaming. Not address unreliable broadcast case

  • Nigel Davis Streaming: Construct the Streaming Spec model (potential for next week call)

There are prior work in TMF that we may gain benefit on (Spec alternatives / requires / excludes / refinement (substitution, division, fusion))

  • Nigel Davis Streaming: Security concern. Before attach to the ControlPort, also determine what the client can do. Enforce at the port. Security of content during transferring.
  • Nigel Davis Streaming: Authenticity of the record (identifies of the creator of the record) RFC5848 (this is just an example of implementation). Both directions (e.g., backpress message from the client, request of replay)
  • Nigel Davis Streaming: Authenticity of "back pressure/replay" requests from client

Actions moved to Task 65 - Streaming

1.2.b

1.3.a

1.3.b

  • View / Context TAPI implications
ND30 mins

oimt2020.ND.006-ViewContext.pptx

3 things (types of activity/performer): Exposing (reporter), mapping (mapper), setting up of mapping (mapping creator. It is driven by the intent)

  • Nigel Davis View/Context: Use case to explore: VN, may have alternative views (two ExposureContext), e.g. for different SLA requirements

Separate namespaces for server context, local context, client context

What represents computation capability? PC? CASC? else? Use Spec approach,

Representation of Library of Spec ? 

  • Nigel Davis View/Context: Make a proposal for the representation of Library of Spec

What do we do with the actual calculation? Who own the view (ExposureContext - EC)?

  • Nigel Davis View/Context: Work on EC
  • Jonathan Sadler View/Context: Send link to ND
  • Hing-Kam Lam View/Context: Making the latest ITU-T arch (G.7701 and G.7702) from the corresponding activity available to OIMT
    • Published in-force:
      • G.7701 Common control aspects 
      • G.7702 Architecture for SDN control of transport networks
    • Draft revision/amendment in-progress: Will request Q12/14 to liaise by end of the its correspondence activities.

In the Spec: View change. Deep control (e.g., packet counts) that causes view change.

The Spec defines the capability, e.g., spot light, flash light, to activate/deactivate, 

Actions moved to Task 11 - View/Context

2.1.a
  • Use of Job Task to improve Ops Patterns
ND30 mins

Note: Moved from 1.3b as this flows better with View/Context and Streaming

oimt2020.ND.007-JobTask.pptx

Note: Moved from 1.3.a to 2.1.a

  • Malcolm Betts : To study the boundaries of Job/Task, ControlConstruct, PC, CASC (algorithmic). Consider path computation as an example.

Work flow of Task (T57)

  • Chris Hartley To propose a general task model, including temporal expression

Aim for extending TAPI.

2.1.b
  • Simplified Spec vs Generalized Spec
ND30 minsT56 vs T37

Note: Moved from 1.3a

oimt2020.ND.008-Spec.pptx

Note: Moved from 1.3.b to 2.1.b

In TAPI, the photonic layer FC spec is very complex. It need to be understandable by a smart controller. 

Pursuing Simplified spec model for TAPI usage, including LTP Sped and FcSpec. It will then be taken to augment the TAPI model.

Agreed with the proposal (slide #5)

  • Nigel Davis To build the LTP Spec and FcSpec usages for v1.5.

Actions moved to Task 56 - Simplified Spec Model

2.2
  • Discuss approaches for Equipment
ND, KL60 minsStrategy of approach

oimt2020.ND.004-PrepForVirtualFaceToFace.pptx

  • Nigel Davis Hing-Kam Lam: To revisit the equipment property model. To change from composite to decorate. This is the "Simple" one.

Once identify a property (e.g., temperature), then decorate it at specifc level of the equipment decomposition (e.g., at the CPU of the equipment). 

Consider have a dot A document (.A.-)

General model of property or just focus on physical equipment property.

2.3
  • Model Structure Update
ND30 minsClose off on strategies

Note: Moved from 1.2a as this fits better with Constraint Domain Split (ND)

Note: Moved from 2.1.a to 2.3.a 

oimt2020.ND.009-ModelRestructuring.pptx

  • Nigel Davis: To urgently discuss with TAPI on application of UUID as agreed in the Sydney meeting
  • Nigel Davis Hing-Kam Lam: To look at addressing the tooling funding challenge. 

Tooling is more than just uml2yang, but implementation forms, including injecting identifier, etc., 

  • All: To make the conclusion on LtpPort at the Friday summary session.
  • Nigel Davis: Talk to Martin on Papyrus model migration (for LtpPort) 
  • Nigel Davis Discuss with the TAPI and Wireless transport: Moving away from simple packaging, e.g., object class, datatype, to a much deeper structure. This will not work with the uml2yang tool. Therefore the model will need to be refactored into object/datatype for uml2yang tool.

Friday:

  • Nigel Davis: To work through the model for v1.5: repackaging.
  • Nigel Davis: v2.0 - Resolve loop, reverse association 
3.1
  • Ethernet Examples
KL60 minsNeed to look at IEEE, ITU-T, TAPI and OIMT aspects

oimt2020.KL.004_vMtg-2020.04_T63-EthernetExamples.pptx

Clarified that PoE injection/extraction at Access Port: This is just on the Physical layer (to the left)

  • To show Q-in-Q bridge (#26) in ITU-T style diagram (or MEF-like)
  • Hing-Kam Lam Revisit the scope of Task 63. Update the description. Need to discuss what we want to achieve: Just additional examples in A.6, or Modeling of IEEE bridges in LTP artifact form (Ltp, Ctp, Ttp, ...)
  • Andrea Mazzini : To provide specific Ethernet example needs
3.2
  • Security 
JS60 minsAuthorization/Access Control

Task: Extend the Core (party, task/job) with addition attribute for security policy. 

Decision of N-ACM vs XACML is a separate question.

http://docs.oasis-open.org/xacml/3.0/errata01/os/xacml-3.0-core-spec-errata01-os-complete.html

Application: Apply this framework for virtual network operation (configuration).  

  • all: Clarify the security work item for authorization (in addition to authentication).
3.3a
  • Intent
ND30 mins

Note: Moved from 2.3 to 3.3

oimt2020.ND.011-Intention.pptx

Temporality - changes 

  • Nigel Davis Chris Hartley: Explore mechanism converting explicit property sentence into constrainable constrain statement.

This is related to security. 

4.1.a
  • Constraint Domain Split
ND30 minsUpdate on practicality of this - does it solve the problem ?
Update strategy - agree to move forward to it ?

Note: Moved from 2.1.b to 2.3.b 

Note: Moved from 2.3.b

Note: Moved from 3.3.b

CDB defines a boundary

CRSP enforces the boundary

4.1.b
  • Summary
KL60 minsRe-visit Bubble chart slide showing work items and their relationship / planned order

Note: Moved from 3.3 to Friday 4.1

Bubble chart

Packaging, association navigability changes

Attendees




  • No labels