Versions Compared

Old Version 72

changes.mady.by.user Stephen terrill

Saved on

compared with

New Version Current

changes.mady.by.user Kenny Paul

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

This wiki page is explicitly set up as a scratch-pad of feedback that the ONAP TSC Members will use to identify, and collate initial feedback to ONAP project proposals as we all prepare for the first set of Project Creation Reviews starting on June 8th and 9th 2107.

If you are not a member of the ONAP Technical Steering Committee (TSC), please do not post your comments here.  If you have comments you would like to make on a project proposal, please do it as a comment on the Project's Proposal Page.  This page is reserved exclusively for use by TSC Members.

Guidelines for TSC Members

Technical Steering Committee Members, please provide your name/initials at the start of your comments.  The proposals below are organized by focus area as we discussed previously.

(Mazin)

Thank you for all your hard work.

As a general comment to all projects. Please be very clear on your scope. Identify on who the lead is and committers.

Committers need to be individuals who have knowledge of the code and have contributed to it. Also state whether

you plan to contribute seed code that you have and the maturity of that code. Risk assessment and clarity are critical

dimensions to TSC approvals.

Orchestration & Controllers

APPC Project Proposal (5/12/17

(Phil Robb pjr) SAMPLE COMMENT

(pjr) Sample Comment #2

(Stephen Terrill)

  • While obvious, it would be good to state in the project description and scope that this produces the APP-C module.  E.g. This purpose of this project is to create the Application controller module.
  •  In the project description, it states the "APPC model will be based on ONAP TOSCA, YANG, ...".  I assume this will have a dependency on the modeling project.  Can we call out specifically what this project will contribute with the models vs uses.  i.e. I was looking for whether this project felt it was responsible for a model, in alignment with the modeling project, Or would it use the models defined by the modeling project?
  • The project description could go into more of what the project is about.  e.g. To produce an application controller that : ...."
  • In the scope, where it states that it "provides Generic LCM commands" it would be good to work in that this is an API that is the responsibility of this project. 
  • There is some text such as "Manage the VNF operational state including Blocking, Sequencing and  Session Throttling" where the project is not really doing, but the model the project produces does, small but could be clearer.  e.g. "APP-C will: bullet points.
  • In the text for "how does this align with external standards/specification", I thought it also used YANG for the components of the southbound? (question, not a statement).
  • General: Is there dependencies on the common controller framework that hasn't been called out?  Either state the dependencies, or state how that is to be handled in the scope.
  • General: There is the discussion around the alignment or relation to the VF-C, however I do agree that leaving this out for now until this is clearer in the architecture is fine.
  • I think it would be good to call out the interfaces defined and used by the project.  Note as there are interfaces common to all controllers, maybe the definition of the interfaces could be under the common controller framework project??? note: we need to ensure that this is done in such a way that the capabilities of one controller, or release, does not hold back the capabilities of other controlers.
  • My view on the resources, committers is that this is ok (4 committers, 3 companies, a number of interested resources).

(Zhaoxing Meng)

  • From the view of overall architecture, APP-C is used for VNF configuration. But the scope of APP-C proposal includes a lot of functionalities about the VNF Lifecycle Management. We need to make it clear how to do VNF lifecycle management from Architecture level first, then we can take a look at the scope of the APP-C to see if it align with the architecture.

(Chris Donley CJD)

  • As noted by previous commenters, the project description and scope are unclear.  It would help if you explicitly describe the problem you are solving and then your approach.
  • Would you please provide more information on your support (or planned support) for TOSCA and ETSI NFV MANO?

VF-C: Virtual Function Controller (5/16/17)

(Stephen)

  •  Scope: Skeleton text of "provide the functionality" can be removed and replaced with "This project provides a VF-C that":
  • I think it would be good to call out the interfaces defined and used by the project.  Note as there are interfaces common to all controllers, maybe the definition of the interfaces could be under the common controller framework project??? note: we need to ensure that this is done in such a way that the capabilities of one controller, or release, does not hold back the capabilities of other controllers.
  • As per the APP-C project, we need to state what models this is responsible for defining, vs what it uses.  coordination with model project and app-c project. 
  • JIRA etc needed.
  • It would be good to state either dependencies on the common controller framework, and or how that will be worked through.
  • The committers need to be called out. (resourcing seems OK).

(CJD)

  • Will you please fill out the key project facts section at the bottom?  It will make it easier for LF to set up the repos and mailing list.
  • It would help to include more information in the description.  

SDN-C (5/12/17)

(Stephen)

  •  Scope: It would be good to clarify which interfaces are defined by this project and which it uses.  Note: proposal to have the common controller framework defining the interfaces?  (can be discussed).
  • Scope: State which models this is dependant on, and which it is "responsible for" - coordinate with modeling project.
  • General: It would be good to state which parts of the common controller framework this will use, or at least how that will be aligned.
  • Resources OK.
  • It seems a little high on the committers.  There are several companies, that is great.  Would it make sense to have one per company? 

(CJD)

  • Would you please clarify the scope of the project?  You say global controller that manages network resources.  What does that mean?
    • Is it limited to data center SDN overlays (as ECOMP SDN-C is today)?
    • Are you considering WAN connections? How about underlays? Are they in scope here, or are you envisioning someone would start a separate project in the future to address these items?

Multi VIM/Cloud (5/11/17)

(Stephen)

  • Scope: It would be good to be clear on the modules this produces and APIs it defines.  It feels like it is a new module (as one can read the infra controller as being openstack, and this seems to sit in between).
    • Related: This has an impact on SO and controllers as they assume openstack today.   This will be a new interface to be defined (by this project I assume) that needs to be implemented by SO,VF-C, APP-C.  Is this handshaken with them (comments suggest this is the case, but its not clear to me).
  •  Scope: Closed loop remediation: Would be good to be clear on the relationship with DCAE here.  Is this to deliver a module to be used by DCAE, or should that part be part of DCAE?  Keep in mind that the module relates to a repo that has a set of committers

Scope: "should align with common controller framework" - any chance to be more specific as to what is being considered here.

(Lingli Deng)

  • Need more specific scope clarification on dependency/impact to DCAE in how FCAPS data would be collected and abstracted via multi-cloud and fed into DCAE in a infrastructure-agnostic manner to address the usecase requirement for cross-layer fault correlation.

(CJD) I'm confused about the closed loop remediation, as well.  

Service Orchestrator (5/14/17)

(Alla, Andrei)

  •  Closed Loop management and orchestration: applying Event/Model/Policy approach causing service re-plan

(Stephen)

  • Scope is rather clear.  It would be good to state which APIs the project feels it is responsible for defining (it states later which it feels it uses).
  •  JIRA prefix: I suggest we drop MSO now as I understood its decided as SO.

(CJD)

  • All participants are listed as committers (34).  Please review and decide who is a committer and who is a contributor.

ONAP Operations Manager (5/10/17)

(Stephen)

  • I was unsure whether this is two projects or one (ONAP operations manager and ONAP operations manager / ONAP on containers).  I am not sure how to differentiate the deliverables from the projects.
  •  While it may be obvious, it would be good to call out the modules being produced
  • Does the project feel that there are APIs that it is responsible for?
  • The committer list is rather long, 1-2 per company?

(Jason Hunt)

  • Good proposal.  This is an important project to help with the consumability of ONAP across users as well as ensuring efficiency of the overall ONAP development process.  You might want to list projects that would benefit from ONAP Operations Manager, such as the virtual and physical lab projects.
  • There is overlap with the Microservices Bus proposal in the health checking of components and service registry.  There should be a single project that can check the health of not only the VM/container but the running microservice as well.  I highly recommend leveraging existing open source projects that will do this health checking and registry for you.  Possibilities include: Consul, Eureka, Marathon's service discovery, or Kubernetes service routing… or the recently announce istio (from Google, IBM, and Lyft).
  • For the components within ONAP Operations Manager, is there a possibility to leverage other ONAP components to do this?  For example, the UI component should be accessible from the main ONAP portal.  Understand that there may be a bit of a bootstrapping issue, of course.
  • For the following portion of the project description, you might want to clarify that this is only for control loop actions on the ONAP components themselves, not the VNFs: “monitor actions that have been taken as part of a control loop (e.g., scale in-out, self-heal)”

(CJD)

  • Would you please clarify your resources list? Who is a committer and who is a contributor?

Testing & Certification

(CJD) General comment for the three related projects (VNF Requirements, VNF SDK, and ICE):  These projects are closely related, and we worked to align the scopes.  The committers for each of the projects are different, which is one of the main reasons we made the split where we did.  VNF Requirements is creating requirements for VNF vendors (documentation), VNF SDK is producing tools (code), and ICE is producing the validation testing program (governance, etc.). The skillsets are different.

VNF Guidelines (Renamed as VNF Requirements)

(Lingli Deng)

I suppose it is now renamed as VNF Requirements, and hence the following comments goes to the proposal at:  https://wiki.onap.org/display/DW/VNF+Requirements

  • Potential Overlap with VNF SDK for VNF validation tooling or testing framework.
  • Need clarificatin with the overlap between modeling and this project regarding the VNF template (HEAT and TOSCA).

ICE

(Alla, Andrei)

  • VNF package validation, VNF testing based on agreed KPI's;
  • Supporting ETSI NFV standards and TOSCA specifications;
  • VNF store/market place;
  • VNF certification;
  • Leveraging SDC platform and SDC/ONAP Portal

(Jason Hunt)

  • I believe I understand the difference between ICE and VNF SDK, as it appears that VNF SDK will provide all the tooling to implement the validation program, but the ICE/VNF Validation Program will establish the community and processes for validation.  Is this correct?  Because of the tight synergies of these projects, would you consider merging them?  Perhaps VNF SDK and VNF Requirements could be subprojects under the VNF Validation Program?
    • (CJD) The skillsets (and interests) of the committers are different.  Some people have experience with code, others with writing guidelines, and others with developing certification/compliance programs.  While there will be a relationship between the projects, we think that this separation makes sense from a committer and governance perspective.

(Roberto Kung) 

  • VNF SDK and ICE should be merged. VNF-SDK starts almost from scratch, so it would be easy. Some links to be made do OPNFV (Dovetail, Program CVP).
  • Will depends very much on technical decisions from other projects that may impact VNF guidelines.
    • (CJD) VNF SDK is part of the OPEN-O Mercury release (not from scratch).  Note that we made changes from the original project proposals - code development (tools) will occur in VNF SDK and validation program development (governance) will occur in ICE.

(Lingli Deng)

  • +1 to VNF SDK and ICE should be merged as a unified tooling project. 
    • (CJD) that's the VNF SDK proposal. ICE has been updated to focus on the validation program governance.
  • Either way, the tooling for packaging and validation implemented by these tooling projects should be fully complying with the unified modeling (modeling project) and guidelines/specifications procedural requirements (VNF requirements) so that they are not diverging from each other.
  • It seems that in the revised project proposal focuses more on the program and test cases, but still contains tools/testing framework, which has potential overlap with VNF SDK, e.g. what tools to be used when conducting VNF lifecycle test for validation, etc..

(Ranny Haiby)

  • There need to be multiple levels of certification. ONAP compliance is more complicated than a binary yes/no. For example, a VNF may be ready for instantiation by ONAP, but not ready for monitoring by DCAE, or life-cycle-management by the APP-C. So the project should clearly state what are the aspects of certification, and what certification "packages" may be available. 
  • -1 to the VNF-SDK/ICE merge proposal. The two projects will likely have different repositories, skill set, etc. The fact that the SDK serves ICE does not necessarily mean they must be merged. ICE can use additional tools than the VNF-SDK, and the VNF-SDK may be used outside the scope of ICE

(Mazin Gilbert)

This comment is for both ICE (verification) and VNF-SDK. Please ensure there is no overlap in the functions. You should also discuss potential technology (tool) overlap.The two components need to work in a streamlined approach. Also connect with the SDC project to ensure both components intgerate into that architecture. The exclusive purpose of this wiki page was to consolidate all TSC feedback during the original 30+ ONAP project creation reviews that occurred between June 8th and 9th 2107. 
It is retained for historical purposes only.

General Feedback for All Projects

(Mazin)

Thank you for all your hard work.

As a general comment to all projects. Please be very clear on your scope. Identify on who the lead is and committers.

Committers need to be individuals who have knowledge of the code and have contributed to it. Also state whether

you plan to contribute seed code that you have and the maturity of that code. Risk assessment and clarity are critical

dimensions to TSC approvals.


(Phil Robb)

Similar to Mazin's comment, please choose your initial set of Committers carefully.  As a general rule, you should not have more committers than contributors.  Please refer to the email I sent on this topic for more information.

Also, for your committers in particular, make sure you provide their gerrit IDs.  The Linux Foundation Infrastructure Team will need this information to give them committer access rights within the build system.

Orchestration & Controllers

APPC Project Proposal (5/12/17

(Stephen Terrill)

  • While obvious, it would be good to state in the project description and scope that this produces the APP-C module.  E.g. This purpose of this project is to create the Application controller module.
  •  In the project description, it states the "APPC model will be based on ONAP TOSCA, YANG, ...".  I assume this will have a dependency on the modeling project.  Can we call out specifically what this project will contribute with the models vs uses.  i.e. I was looking for whether this project felt it was responsible for a model, in alignment with the modeling project, Or would it use the models defined by the modeling project?
  • The project description could go into more of what the project is about.  e.g. To produce an application controller that : ...."
  • In the scope, where it states that it "provides Generic LCM commands" it would be good to work in that this is an API that is the responsibility of this project. 
  • There is some text such as "Manage the VNF operational state including Blocking, Sequencing and  Session Throttling" where the project is not really doing, but the model the project produces does, small but could be clearer.  e.g. "APP-C will: bullet points.
  • In the text for "how does this align with external standards/specification", I thought it also used YANG for the components of the southbound? (question, not a statement).
  • General: Is there dependencies on the common controller framework that hasn't been called out?  Either state the dependencies, or state how that is to be handled in the scope.
  • General: There is the discussion around the alignment or relation to the VF-C, however I do agree that leaving this out for now until this is clearer in the architecture is fine.
  • I think it would be good to call out the interfaces defined and used by the project.  Note as there are interfaces common to all controllers, maybe the definition of the interfaces could be under the common controller framework project??? note: we need to ensure that this is done in such a way that the capabilities of one controller, or release, does not hold back the capabilities of other controlers.
  • My view on the resources, committers is that this is ok (4 committers, 3 companies, a number of interested resources).

(Zhaoxing Meng)

  • From the view of overall architecture, APP-C is used for VNF configuration. But the scope of APP-C proposal includes a lot of functionalities about the VNF Lifecycle Management. We need to make it clear how to do VNF lifecycle management from Architecture level first, then we can take a look at the scope of the APP-C to see if it align with the architecture.

(Chris Donley CJD)

  • As noted by previous commenters, the project description and scope are unclear.  It would help if you explicitly describe the problem you are solving and then your approach.
  • Would you please provide more information on your support (or planned support) for TOSCA and ETSI NFV MANO?

VF-C: Virtual Function Controller (5/16/17)

(Stephen)

  •  Scope: Skeleton text of "provide the functionality" can be removed and replaced with "This project provides a VF-C that":
  • I think it would be good to call out the interfaces defined and used by the project.  Note as there are interfaces common to all controllers, maybe the definition of the interfaces could be under the common controller framework project??? note: we need to ensure that this is done in such a way that the capabilities of one controller, or release, does not hold back the capabilities of other controllers.
  • As per the APP-C project, we need to state what models this is responsible for defining, vs what it uses.  coordination with model project and app-c project. 
  • JIRA etc needed.
  • It would be good to state either dependencies on the common controller framework, and or how that will be worked through.
  • The committers need to be called out. (resourcing seems OK).

(CJD)

  • Will you please fill out the key project facts section at the bottom?  It will make it easier for LF to set up the repos and mailing list.
  • It would help to include more information in the description.  

SDN-C (5/12/17)

(Stephen)

  •  Scope: It would be good to clarify which interfaces are defined by this project and which it uses.  Note: proposal to have the common controller framework defining the interfaces?  (can be discussed).
  • Scope: State which models this is dependant on, and which it is "responsible for" - coordinate with modeling project.
  • General: It would be good to state which parts of the common controller framework this will use, or at least how that will be aligned.
  • Resources OK.
  • It seems a little high on the committers.  There are several companies, that is great.  Would it make sense to have one per company? 

(CJD)

  • Would you please clarify the scope of the project?  You say global controller that manages network resources.  What does that mean?
    • Is it limited to data center SDN overlays (as ECOMP SDN-C is today)?
    • Are you considering WAN connections? How about underlays? Are they in scope here, or are you envisioning someone would start a separate project in the future to address these items?

Multi VIM/Cloud (5/11/17)

(Stephen)

  • Scope: It would be good to be clear on the modules this produces and APIs it defines.  It feels like it is a new module (as one can read the infra controller as being openstack, and this seems to sit in between).
    • Related: This has an impact on SO and controllers as they assume openstack today.   This will be a new interface to be defined (by this project I assume) that needs to be implemented by SO,VF-C, APP-C.  Is this handshaken with them (comments suggest this is the case, but its not clear to me).
  •  Scope: Closed loop remediation: Would be good to be clear on the relationship with DCAE here.  Is this to deliver a module to be used by DCAE, or should that part be part of DCAE?  Keep in mind that the module relates to a repo that has a set of committers

Scope: "should align with common controller framework" - any chance to be more specific as to what is being considered here.

(Lingli Deng)

  • Need more specific scope clarification on dependency/impact to DCAE in how FCAPS data would be collected and abstracted via multi-cloud and fed into DCAE in a infrastructure-agnostic manner to address the usecase requirement for cross-layer fault correlation.

(CJD) I'm confused about the closed loop remediation, as well.  

Service Orchestrator (5/14/17)

(Alla, Andrei)

  •  Closed Loop management and orchestration: applying Event/Model/Policy approach causing service re-plan

(Stephen)

  • Scope is rather clear.  It would be good to state which APIs the project feels it is responsible for defining (it states later which it feels it uses).
  •  JIRA prefix: I suggest we drop MSO now as I understood its decided as SO.


(CJD)

  • All participants are listed as committers (34).  Please review and decide who is a committer and who is a contributor.


ONAP Operations Manager (5/10/17)

(Stephen)

  • I was unsure whether this is two projects or one (ONAP operations manager and ONAP operations manager / ONAP on containers).  I am not sure how to differentiate the deliverables from the projects.
  •  While it may be obvious, it would be good to call out the modules being produced
  • Does the project feel that there are APIs that it is responsible for?
  • The committer list is rather long, 1-2 per company?

(Jason Hunt)

  • Good proposal.  This is an important project to help with the consumability of ONAP across users as well as ensuring efficiency of the overall ONAP development process.  You might want to list projects that would benefit from ONAP Operations Manager, such as the virtual and physical lab projects.
  • There is overlap with the Microservices Bus proposal in the health checking of components and service registry.  There should be a single project that can check the health of not only the VM/container but the running microservice as well.  I highly recommend leveraging existing open source projects that will do this health checking and registry for you.  Possibilities include: Consul, Eureka, Marathon's service discovery, or Kubernetes service routing… or the recently announce istio (from Google, IBM, and Lyft).
  • For the components within ONAP Operations Manager, is there a possibility to leverage other ONAP components to do this?  For example, the UI component should be accessible from the main ONAP portal.  Understand that there may be a bit of a bootstrapping issue, of course.
  • For the following portion of the project description, you might want to clarify that this is only for control loop actions on the ONAP components themselves, not the VNFs: “monitor actions that have been taken as part of a control loop (e.g., scale in-out, self-heal)”

(CJD)

  • Would you please clarify your resources list? Who is a committer and who is a contributor?

Testing & Certification

(CJD) General comment for the three related projects (VNF Requirements, VNF SDK, and ICE):  These projects are closely related, and we worked to align the scopes.  The committers for each of the projects are different, which is one of the main reasons we made the split where we did.  VNF Requirements is creating requirements for VNF vendors (documentation), VNF SDK is producing tools (code), and ICE is producing the validation testing program (governance, etc.). The skillsets are different.

VNF Guidelines (Renamed as VNF Requirements)

(Lingli Deng)

I suppose it is now renamed as VNF Requirements, and hence the following comments goes to the proposal at:  https://lf-onap.atlassian.net/wiki/display/DW/VNF+Requirements

  • Potential Overlap with VNF SDK for VNF validation tooling or testing framework.
  • Need clarificatin with the overlap between modeling and this project regarding the VNF template (HEAT and TOSCA).

ICE

(Alla, Andrei)

  • VNF package validation, VNF testing based on agreed KPI's;
  • Supporting ETSI NFV standards and TOSCA specifications;
  • VNF store/market place;
  • VNF certification;
  • Leveraging SDC platform and SDC/ONAP Portal

(Jason Hunt)

  • I believe I understand the difference between ICE and VNF SDK, as it appears that VNF SDK will provide all the tooling to implement the validation program, but the ICE/VNF Validation Program will establish the community and processes for validation.  Is this correct?  Because of the tight synergies of these projects, would you consider merging them?  Perhaps VNF SDK and VNF Requirements could be subprojects under the VNF Validation Program?
    • (CJD) The skillsets (and interests) of the committers are different.  Some people have experience with code, others with writing guidelines, and others with developing certification/compliance programs.  While there will be a relationship between the projects, we think that this separation makes sense from a committer and governance perspective.
    • (ES) To add a bit to what Chris added. We have discussed in depth around the division between the VNF SDK & Tooling and ICE project. There may be some overlap between the projects as far as tooling but the close work between the projects will allow us to make sure it is ironed out during the course of the projects. The long-term plan is to leverage the VNF SDK & Tooling and VNF Requirements as basis for the VNF Validation Program. There may be items inside the VNF Validation Program which will not be covered by the VNF SDK & Tooling which would then mean it would be supplemented under the VNF Validation Program umbrella. Additionally any tooling which exists in the current ICE implementation will be contributed into the VNF SDK & Tooling project.   

(Roberto Kung) 

  • VNF SDK and ICE should be merged. VNF-SDK starts almost from scratch, so it would be easy. Some links to be made do OPNFV (Dovetail, Program CVP).
  • Will depends very much on technical decisions from other projects that may impact VNF guidelines.
    • (CJD) VNF SDK is part of the OPEN-O Mercury release (not from scratch).  Note that we made changes from the original project proposals - code development (tools) will occur in VNF SDK and validation program development (governance) will occur in ICE.

(Lingli Deng)

  • +1 to VNF SDK and ICE should be merged as a unified tooling project. 
    • (CJD) that's the VNF SDK proposal. ICE has been updated to focus on the validation program governance.
    • (ES) The division of the projects have been discussed between the teams. We are aware of the close linkage between these two projects (as well as other relevant projects) yet believe it will be ironed out during the course of the projects. 
  • Either way, the tooling for packaging and validation implemented by these tooling projects should be fully complying with the unified modeling (modeling project) and guidelines/specifications procedural requirements (VNF requirements) so that they are not diverging from each other.
    • (ES) Any validation will be based the combined efforts of the platform teams including modeling. The initial scope of the VNF Validation Program aims to focus on the VNF Package for on-boarding yet the process defined is not limited to that only.  
  • It seems that in the revised project proposal focuses more on the program and test cases, but still contains tools/testing framework, which has potential overlap with VNF SDK, e.g. what tools to be used when conducting VNF lifecycle test for validation, etc..
    • (ES) There may be an initial overlap of the code made available in the repos yet the long-term plan is to leverage the VNF SDK & Tooling and VNF Requirements as basis for the VNF Validation Program. It is likely necessary to complement that tooling with validation process specific items. The details around the delineation will be ironed out during the project with the continued close contact between the various teams involved.  

(Ranny Haiby)

  • There need to be multiple levels of certification. ONAP compliance is more complicated than a binary yes/no. For example, a VNF may be ready for instantiation by ONAP, but not ready for monitoring by DCAE, or life-cycle-management by the APP-C. So the project should clearly state what are the aspects of certification, and what certification "packages" may be available. 
    • (ES) Agree that there are multiple levels of validation/certification. The initial focus is to establish the governance of the project and define a process allowing anyone to obtain a ONAP Compatible Label for their VNF. We will also focus on making sure we establish process to extend the scope of the validation to cover additional scope and aspects. For the initial release the main focus is on the VNF Package content and validity though. 
  • -1 to the VNF-SDK/ICE merge proposal. The two projects will likely have different repositories, skill set, etc. The fact that the SDK serves ICE does not necessarily mean they must be merged. ICE can use additional tools than the VNF-SDK, and the VNF-SDK may be used outside the scope of ICE
    • (ES) Agree that even though there is a close relationship yet are by their nature very different. I do actually believe it to be more than likely we will leverage additional tools than what will be in scope for the VNF SDK & Tooling yet it is important to minimize any such tools to the extent possible.

(Mazin Gilbert)

This comment is for both ICE (verification) and VNF-SDK. Please ensure there is no overlap in the functions. You should also discuss potential technology (tool) overlap.The two components need to work in a streamlined approach. Also connect with the SDC project to ensure both components intgerate into that architecture.

  • (ES) I should have covered most of this in my comments above. The VNF SDK & Tooling, VNF Requirements, SDC and ICE project teams have discussed in depth the necessary close collaboration between the projects and the need to manage any potential overlap. The long-term plan is to leverage the VNF SDK & Tooling and VNF Requirements as basis for the VNF Validation Program. One key item as part of the first release of the VNF Validation Program is to define that plan. There may be items inside the VNF Validation Program which will not be covered by the VNF SDK & Tooling which would then mean it would be supplemented under the VNF Validation Program umbrella. Any tooling which exists in the current ICE implementation will additionally be contributed into the VNF SDK & Tooling project.   

(Chris) We discussed this at length between the VNF SDK/ICE teams.  ICE/Verification will focus on the verification program.  VNF SDK will focus on the tools (including the ICE tools) and will align with SDC.  

VNF-SDK

(Alla, Andrei)

  • VNF package onboarding;
  • Supporting ETSI NFV standards and TOSCA specifications;
  • Leveraging SDC platform and SDC/ONAP Portal

(Jason Hunt)

  • I believe I understand the difference between ICE and VNF SDK, as it appears that VNF SDK will provide all the tooling to implement the validation program, but the ICE/VNF Validation Program will establish the community and processes for validation.  Is this correct?  Because of the tight synergies of these projects, would you consider merging them?  Perhaps VNF SDK and VNF Requirements could be subprojects under the VNF Validation Program?  Or are there other uses of VNF SDK outside of the validation program?
  • Can you clarify how this tooling will be provided?  Will it be another running component of ONAP, accessible from the ONAP portal?
  • Can you clarify the role of the network function repository?  Is the thought for this to be run separately by a third party?  If not, and it is part of an ONAP installation, how is it different from the existing catalog of onboarded functions? 

...

  • This comment is for both ICE (verification) and VNF-SDK. Please ensure there is no overlap in the functions. You should also discuss potential technology (tool) overlap.The two components need to work in a streamlined approach. Also connect with the SDC project to ensure both components intgerate into that architecture. 

Integration (5/11/2017)

Reference VNFs Project (now part of the Integration Project)Analysis

...

we need to clarify how it works with OPNFV CI/CD or PF (should be easy as some actors Huawei/Orange/ATT … are in both). May be a big project (but it builds on OPEN-O experience, and on what will be brought by OpenLab)brought by OpenLab)

Helen Chen: We are trying to leverage OPNFV lab space for ONAP R1 CI/CD testing, especially from developer point of view. In the longer term, we are looking for a more harmonic way to collaborate with them.

I agree that this project is big, but all aspects are tightly related. We have subgroup who are responsible and focusing on each category, and collaborate with the rest of Integration team.

(Ranny Haiby)

  • It is not immediately clear from the project definitions what types of labs will be required. It is implied by the first diagram, and also discussed by Catherine in some email threads. Perhaps it would be best to explicitly describe what types of labs will be required by the integration project.
    Helen Chen] Please refer to this doc: Test requirement from developer point of view

(Mazin Gilbert)

  • Let's consider separate proposals for the Labs versus Integration. Labs should define the minimal assets required to stand up a lab (Physical, Power, Networking, People, Software)Networking, People, Software).
    Helen Chen: Agreed, even thought Integration project and Labs have a lot of overlaps regarding to CI / CD, Use case deployment, configuration, service template design,  etc.,  and we may end with the same group of people.  These two projects will work closely: Integration will use the Labs built by the community. So far, basic pod definition has been proposed. Based on approved use cases, we will work with the team to finalize the minimal lab requirements.

Analysis

DCAE (5/11/17)

(Roberto Kung)

...

  • We might want a portal for DCAE to enable the user to have a systematic view of the information gathered. For example, considering the requirement for performance monitory, we need a UI interface to ops rep to monitor the performance stats.


Holmes (5/11/17)-


(Roberto Kung)

Holmes should be looked with Clamp or/and Policy, mainly policy (with introduction of engines and so on). May be a split is needed (analytics – alarm aggregation, filtering, correlation in DCAE analytics microservices / policy design RCA in policy). May not be high priority for R1 (not needed for our use cases). But it is useful to show intents for following releases

...

  • This project should be split and combine with DCAE (for the correlation engine), Policy engine (for Drools), and CLAMP (for designing the open loop).
  • (Lingli Deng) What about the portal demonstrating the alarms gathered, and correleation made? Would DCAE be providing a portal for that?

Common Platform & Utilities

ONAP Extensibility (5/16/17)

(Zhaoxing)

  • The suggested guidelines and approaches from extensibility project should be discussed among projects and agreed in the TSC before enforcement across projects.

...

This project targets to resolve an important problem in ONAP development and need cooperation with almost all projects. Need to define process/methodology/checkpoints to form and enforce the recommended patterns.

Authentication and Authorization Framework (5/16/17)

(Zhaoxing)

  • Could the project proposal clarify how to enforce the auth of asyn message publishing/subscription and the sync REST requests separately?

  • Will this project enforce access control between the components inside the ONAP system, for example, the communication between SO and A&AI?

  • Will this project implement centralized auth for all the ONAP components, or user need to log into every component? If that's the expectation, AAF may need to discuss with MSB team about the possibility of the AAF MSB integration approach for centralized auth, which is listed in the MSB use cases.

  • Is there any internal or external dependencies for this project?

  • No enough participant companies, according to the proposal framework, a proposal should have at least 3-4 companies involved to provide diversity

  • Lack of link to seed code

...

Need to consider authentication federation with other parts like underlying OpenStack Cloud. Better if more company diversity.

Microservices Bus(5/12/17)

(Zhaoxing)

  • Could the project proposal clarify more about its relationship with AAF and Dmaap?

...

Clarifify the call flow with other related projects will help the overall design.

Common Controller Framework (5/12/17)

(Zhaoxing)

  • I would suggest renaming this project to Common Controller SDK because it provides building blocks for controllers.

  • Will this project provide source codes or binaries(libs) for controller projects?

  • Could this project proposal clarify what exactly is "Common microservice / VF lifecycle management"? It seems conflict with OOM and SO/VF-C/APP-C.

  • Could this project proposal clarify what exactly is "Common model management"? It seems to conflict with modeling project.

  • Lack of link to seed code

...

This project seems to want to resolve code reuse problem among all the controllers. Need more clarification on the concrete scope. High level terms are listed in the scope section like Common health checks. Need to provide more fine-granularity design for more understanding of overlap with others. The list of dependencies with other open source projects need more explanation too.

DMaaP – Data movement as a platform (5/16/17)

(Zhaoxing)

  • Lack of link to seed code

...

This project targets to provide data filtering/processing/transportation support by leverage other open sources. With the big scope, the team need to clarify what is the focus in R1. Also need seed code list and company diversity.

Modeling & Design

Note on this section:

The part under the names is comments received from the individuals.

Where it says "review comments" is the consolidated comments after discussion, following the criteria/skeleton proposed.

Modeling (5/11/17)

Review:

Clarity of Scope:

Reasonable and Well defined Scope:

  • Could the proposal clarify the unified model-driven approach? How are the data models used by AAI, APIs, or blueprints updated?
  • Include in scope: Modeling and Design (concept for multiple projects: CLAMP, SDC, AAI, Network Functions Change Management, External API Framework, External System Registry
  • Include in Scope: Application Modeling (VNF modeling) (e.g. YAML based) for managed VNF's as supported by APP-C that leverages Closed Loop management including application management
  • Small: the text uses OPEN-O and OpenECOMP.  As the project description should survive the first release, perhaps look at wording to avoid referring to those terms as they should be deprecated in ONAP.  e.g. "The project will produce unified and consolidated data models".
  • Under "scope" The reference to R1 may better be removed. It may imply that the data model will be designed to address these use cases only and ignore the rest, which I am sure was not the intention. This information should go into the release planning, not project definition.
  • Question: What about backwards compatibility for ECOMP and Open-O data models? Will there be a new model to replace both? If so, is ONAP expected to support only the unified model? Or support the new as well as the two old ones?

...

  • Could the proposal clarify the unified model-driven approach? How are the data models used by AAI, APIs, or blueprints updated?

(Alla, Andrei)

  • Include in scope: Modeling and Design (concept for multiple projects: CLAMP, SDC, AAI, Network Functions Change Management, External API Framework, External System Registry);
  • Include in Scope: Application Modeling (VNF modeling) (e.g. YAML based) for managed VNF's as supported by APP-C that leverages Closed Loop management including application management

(Stephen)

  • Small: the text uses OPEN-O and OpenECOMP.  As the project description should survive the first release, perhaps look at wording to avoid referring to those terms as they should be deprecated in ONAP.  e.g. "The project will produce unified and consolidated data models".
  • During the release planning, it would be good to have the plan of when to deliver what models for the needs of other projects to use.
  • I understand that the deliverables are both models, as well as code in the form of tools.  Should they be in the same repo, or different repos (question, not a statement). consider separating tooling from models in repos
  • The number of committers is very high, perhaps 1-2 per company
  • Be clear on the tools to be as a deliverable, and the APIs.   e.g. a model passer  . 

...

  • On the R1, please state which projects will use which models.  We want to know how the projects are going to interact with the projects and whether they expect to receive the model or receive coordination and guidance. 

CLAMP (5/11/17)

Review Comments:

 Clarity of Scope:

...

Design close loops to connect the models and actions distributed in different components together, such as metrics definition in VNF model, data collection ? analytics in DCAE,  policy and vnf/ns lifecycle management in SO/VF-C/APP-C.

(Eden Rozin): Design close loops is part of the DCAE-D (i.e. DCAE Designer) module which is part of SDC; Close Loops (CLs) are attached to the Design Object (e.g. VNF, Services) and distributed to runtime components (DCAE).

Run time:

Present the status of the running close loops. 

...

Runtime messaging interface between DCAE and Policy

Service Design & Creation (5/17/17)

Review summary:

Clarity of Scope:

Reasonable and Well defined Scope:

  • Scope seems rather large.  Not clear on what is for the first release, and interactions/dependencies with DCAE, policy, SO, Controllers. .....
  • More detail on what test is
  • could be clear on what deliverables are required to support the use cases.
  • What is the scope of the catalog module

Identification of SW modules and APIs being developed and delivered to other components

    • It would be great to be clear on the modules delivered by this project
    • The connections to other projects should be clarified clearly.
    • Are there APIs that the project feels it is responsible for, or does it use the APIs defined by others?

Follow project and LF guidelines for contributors/committers

  • Perhaps 1-2 committers per company

Identification of dependences and assumptions on other components and open source

...

Intentional and unintentional overlap

...

  • Any relation to ICE?
    (Eden Rozin): Process wise; when a VNF is 'Certified' according to standards, thus it passed ICE, in can be onboarded to SDC for further enrichments as a reusable Asset in Catalog; this process-wise handshake can be automated.

overlap with external opensource efforts

...

  • Could the proposal describe the scope of the catalog module?
    (Eden Rozin): The catalog hold all Design Assets as a reusable building block for Design; it holds: VNFC, VNF, Service, Polices, Monitoring Templates (Close Loops, Open Loops), Orchestration Workflows, APIs and Artifacts.

(Alla, Andrei)

  • SDC should be an umbrella for all Design time parts from other Projects (CLAM, Policy, Converged ICE & VNF SD-K);
    (Eden Rozin): Agree.
  • SDC is a platform for all the modeling and design in ONAP and provides the interactive tool for design and automation for on-boarding.
    (Eden Rozin): Agree.

(Stephen)

  • It would be great to be clear on the modules delivered by this project.
  • The connections to other projects should be clarified clearly.
  • Are there APIs that the project feels it is responsible for, or does it use the APIs defined by others?
  • Perhaps 1-2 committers per company
  • More detail on what test is.
  • Scope seems rather large.  Not clear on what is for the first release, and interactions/dependencies with DCAE, policy, SO, Controllers. ......
  • Not clear what deliverables are required to support the usecases

(Zhaoxing)

  • Could this project proposal clarify how to support the Telcom use case(VoLTE), for example, how to onboarding TOSCA based VNF and use it as building block for service design?
  • Could this project proposal clarify its relationship/dependency with ICE? It seems that the ICE project will provide guideline/process definition for certification and this project will provide tools to support ICE?
    (Eden Rozin): Answered in a previous comment.

Active and Available Inventory (AAI) (5/11/17)


Review Comments:

Clarity of Scope:

...

  • In this scope it would be great to be clear on the modules that are to be delivered in general.
  • In the scope it would be great to clarify whether this project responsible for any APIs to be defined, I assume so.
  •  Perpahs 1-2 committers per company?
  • Editorial.  Referens to MSO, should be SO now.

Network Function Change Management Project Proposal (5/11/17)


Review Comments:

Clarity of Scope:

...

  •  It would be great to be clear on the code module deliverables from this project so we can understand whether there are overlaps in the deliverables with the other projects (e.g. SDC).  i.e. what is planned to go into the repo NFCM.  i.e. from the scope of this, it feels like it should be part of SDC or produces requirements, however I would like to understand the deliverables to get a clearer understanding of that.

External API Framework (5/15/17)


Review Comments:

Clarity of Scope:

...

  •  Scope: Question: Does this provide a SB API definition into ONAP (or use one) and a "plug-in" environment for bringing the external interfaces?
  • Scope: Clarify the code modules that are a deliverable.
  • Scope: On the models, discussion with the modeling project about what are the models covered in this project vs the modeling project would be great.  I can see what the line would be though.  the TMF SID maybe one point to align on (not clear how that would be part of integration)

External System Register (5/14/17)

Review Comments:

Clarity of Scope:

...

Sound technical solution solving a real need.

 


(Stephen)

  • Scope: could be a little clearer by way of example.. 

Policy

Policy Framework Project Proposal (5/11/17)

(Chris)

  • Clarity: Project description and scope are unclear. They were written from the perspective of someone who is familiar with the module in ECOMP, but many of the TSC members are not as familiar and we need more clarification.  Please add more detail about the big picture (problem you're solving and vision).  Also, please clarify what you're delivering (code, requirements, etc.).
  • Overlap: How does this relate to other projects? Are you developing APIs? Are you producing requirements that you expect others to implement (and did they agree)?
  • Risk management: It appears that you have sufficient resources.  We would like to see committers from different companies. It would be helpful to clarify your deliverables - is there a subset that can be delivered in the R1 timeframe?
  • Relevance and prioritization: this is relevant to the ONAP release.

...

  • It would be great to state the APIs that this project feels it is responsible to define
  • If possible, consider a committer from another company as well (this shouldn't block approval).
    • Pamela Dragosh - sure we can consider other committers. Initial thought was that for Release 1, best to have only folks that know the code. But the project is designed to evolve the Policy Framework and address its current limitations, which will require us to expand the list of committers.
  • In the scope there is text on classification of policies.  What is unclear to me is the deliverable from that.  Is it a document? is that in internal deliverable?
    • Pamela Dragosh - I don't think we had enough time to clarify that. My thought is that the project should define the flow for all the types of policy required by ONAP. For example, how it is expressed/captured, when it translated, which component calls the API to create/update it, etc. That would at least involve documentation of that flow. I will start discussion on how to clarify this.
  • Seems to have a good level of resourcing
  • Is there any models that this is dependant on, or expected to define?
    • Pamela Dragosh - Yes. The current Policy seed code is model-driven and we utilize models from the DCAE team in order for the DCAE components to be policy-driven. We have used EMF in the past, but are phasing that out in support of TOSCA/JSON approach. We expect to be the team that defines how policy is expressed if there are any specific models that are going to be used to express policy. (i.e. the outcome from the Modeling Project).

...

  • It might be cleaner to keep all references to Release 1 in the release plan.
  • Could the outcome of the classification process also include guidelines to ONAP developers on when the policy framework should be used (and when not)
  • Could you please clarify the scope? Is a Policy Decision Engine one of the deliverable? Is there one central instance of the engine for the entire ONAP or could each module have its own? If central, how do you determine which technology to use?

Policy Driven VNF Orchestration (5/12/17)

(Chris)

  • We understand this is being rolled into Policy Framework
  • Clarity: The scope is unclear. What exactly are you delivering?
  • Overlap: How does this relate to other projects? Are you producing requirements that you expect others to implement (and did they agree)? Is this feeding into SNIRO?
  • Risk management: It appears that you have sufficient resources.  As part of Policy Framework, this could be a longer-term deliverable. Any delay wouldn't derail R1.
  • Relevance and prioritization: this is relevant to the ONAP release.

...

  • +1 to consider openning up to committers from other companies, to encourage participation and contribution where there is clearly specified commitment for Release 1.

SNIRO Optimization Framework (5/11/17)

(Chris)

  • Clarity: Project description and scope are clear. Please finish filling out the Key Project Facts section.
  • Overlap: How does this relate to CLAMP and Policy Driven VNF Orchestration?
  • Risk management: Is there a subset of your functionality that can be delivered in R1?  We would like to see you refine your committer/contributor lists. It would be good to have committers from different companies, but not everyone on your project should be a committer. 
  • Relevance and prioritization: this is relevant to the ONAP release.

...

  • This is a well written project proposal with important scope for a mature ONAP implementation.
  • The scope is quite aggressive.  Can you clarify what scope will be provided in the seed code and what scope is anticipated for delivery in R1?
  • Have you communicated with the project leads for your project dependencies?  For example, Service Orchestrator or Multi-VIM needing to call to SNIRO for a homing solution.
  • The scope item #7 references ONAP-Controller, which is not a current project proposal.  Should this be the ONAP Operations Manager?

UI & Portals

Portal Platform Project Proposal (5/12/17)

1.    Clarity

a.     Reasonable and well defined scope
It is reasonable to have a common portal framework for various UIs from different backend functional modules of the system. But it is not well specified in the project proposal. In particular, in the scope section, it describes the portal, not the portal framework, which is confusing. Suggest to refine the description in accordance with the proposed portal architecture and clarify its scope in relation with other functional modules or portal pages, in more detail. And identify extra functionality or APIs to be implemented during Release 1, if any. To be more clear on the APIs that project feels it is defining.

...

c.    Sound technical solution for solving a real need
Yes. The seed code is mature and deployed for years of practice.

ONAP Usecase UI Project Proposal (5/15/17)

1.    Clarity

a.     Reasonable and well defined scope
It is reasonable to have UI developed for usecases if the requirements are not met by the existing platform portal. It may be a good idea to have the project include a long-lived project scope while leaving the specific use cases to the R1 release plan. For both use cases - Recommend defining the persona using the UI (is it self service? is it used by some care center? sales center? etc. For the VoLTE use case - What is the business use case here? Who is the target end customer? Is it an enterprise customer of the CSP? Is it an internal customer inside the CSP? For both use cases, what is the required functionality? Service deployment? Maintenance? Performance monitoring? Troubleshooting?

...

c.    Sound technical solution for solving a real need
Note clear.


VID project (5/17/17)

1.    Clarity

a.     Reasonable and well defined scope
Not clear whether or not VID would include a portal, and its separation of responsibility for usecase UI requirements and scenarios.

(Eden Rozin): VID will include a portal for Operations to invoke Instantiation of Services and VNFs and to perform Maintanence (Change Management) activities.

b.    Identification of software modules and APIs being developed and delivered to ONAP and other components.
Yes. Suggest to change MSO into SO.

...

c.    Sound technical solution for solving a real need: Not clear, as no technical solution is described.

ONAP CLI

(Lingli Deng) From the list discussion, I was under the impression that the authors of this proposal decided to postpone it for future releases.  

Documentation & Training

Documentation (5/11/17)

1.    Clarity

a.     Reasonable and well defined scope
Scope is reasonable and follow the common practice of open source projects. However, the deliverables is not yet well specified. And the seed code section points only to the openECOMP documentation, which might be confusing.
Will this project be responsible for the documentation? Suggest to clarify the deliverables and clarify the scope of responsibilities and how to enable collaboration between projects/sub-committees that produce documentation and this project. Examples include: architecture from the Architecture Sub-Committee, models from Modeling project, use cases from potentially the Usecase Sub-Committee, etc.
Will this project be responsible for the automatic documentation tooling? Suggest to clarify the scope of responsibilities and how to enable collaboration between the projects that produce tooling and this project. Examples include: integration, common platform services, etc.

...

c.    Sound technical solution for solving a real need
Yes. The team is evaluating the use of readthedocs.org as way of publishing documents, and the use of swagger.io for API documentation. Need to work closely with the release manager in planning deliverables and tooling provision in time.

ONAP University (5/11/17)

1.    Clarity

a.     Reasonable and well defined scope
It is important for a complicated open source software to provide training material to end users as well as developers to attract more participation and adoption.
However, the scope description can be more clear and distinctive from what is already stated in documentation project proposal, perhaps in terms of the form of deliverables, the people involved, etc.

...

c.    Sound technical solution for solving a real need
Yes. Various forms of trainings are suggested, including Video/Online/Classroom/Hands-on/Blog/Webinar, but no specific deliverable/budget planning involved, need Marketing Committee guidance and commitment in the planning and practice.

Misc.

Open Lab (5/23/17)

Agreed to postphone it for further formal project proposal.


VNF Requirements

Grouped under the category of "Testing and Certification" category above.

...