Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Next »

Date 

 


Attendees

Bob Papa

Yuriy Malakov

s.silvius

Ciaran Johnston

Andy Mayer


Discussion Items

ItemWhoNotes
AgendaBob Papa

The purpose of the meeting was to have each group the has done some sort of ONAP Model Driven/Enabled in the past year to explain their view.

**Meeting was not recorded**

 General DiscussionAll  
  • None of the 3 presentations contradicted any others presentations
  • Everyone was in agreement that we should have a common internal model because;
    • Custom code causes delays and increase cost
    • ONAP components should only process one internal model  
  • Need to preserve on boarding artifacts
  • We should probably look at the SDC Internal model to see what is actually in it
  • Need to take an inventory of what components are using TOSCA model concepts
  • Need to define Gaps
  • 3 other work efforts are taking place related to Model Driven ( CDS , API Documentation Project, SO POS (Driven by TOSCA syntax) )
  • CDS is a project that has implemented some TOSCA syntax processing  
    • Yuriy talked about CDS
    • He presented CDS at Antwerp
    • Link to what Yuriy presented in Antwerp is listed above.
    • Policy is using same concepts
    • Not all people were aware of CDS- will look into learning more about it  
  • API Documentation Project
  • SO POC – processing driven by TOSCA syntax


  • Next Steps
    • Continue meeting every two weeks on Monday
    • Look at the current internal model
    • Define what is are target goal and how to get there without breaking things that work.
    • Define any Gaps.
    • Who else should be part of these discussions?

First review by Samuli from Samsung.

s.silvius

  • Reviewed a couple slide from his CNF presentation (link to deck is posted above)
  • Focused was onboarding process
  • ONAP has made multiple attempts to generalize modeling and orchestration
    • TOSCA
    • ETSI
    • Blueprints , etc…
  • Need to choose a single common modeling language  - TOSCA
  • TOSCA is expressive and extensible
  • Any technology should be converted to an internal TOSCA format
  • Some translators already exist in ONAP

Samuli (Samsung) presentation from Antwerp ….  https://wiki.lfnetworking.org/display/LN/ONS%3AEU+%2719+Technical+Event+Schedule?preview=/19006711/22675630/CNF_Modeling.pptx

Second Review  by CiaranCiaran Johnston
  • Ciaran summarized some slides from his full presentation ( link to full presentation above)
  • A high level view of modeling
  • Focused more on the interoperability between components  / api’s
  • Types of models – capability, extension and user
  • Need better API documentation , version control
  • Moving forward need a catalog  of models
  • Should probably include process models

Ciaran (Ericcson)  presented …   https://wiki.lfnetworking.org/download/attachments/15630468/Modeling%20ONAP%20v0.3.pdf?version=1&modificationDate=1560429554906&api=v2

Third Review …  Andy Mayer

Andy Mayer  
 
  • Andy used the first couple of slide from his deck ..(attached)
  • Discussed model driven versus model enabled
  • Showed the model flow…. Create , Deploy and use Model
  • Described the onboarding flow:
    • Different types (yang, Heat, etc. ) are used as input to Service Designer (SDC)   
    • Each file is used as a building block to be converted by SDC to TOSCA Model (Internal)
    • Preserves some artifacts because not all info can be represented in TOSCA
  • Similar process as described by Simuli’s review
  • <<Model_Enabled_ONAP.pptx>>









  • No labels