Date:
Attendees
Shankaranarayanan Puzhavakath Narayanan
Agenda
- Decision on use of E2EST
- Allotted resources vs alternative approach
- Next actions – modeling for M2, APIs, other open points (S-NSSAI, etc.)
- EXT-API queries
- If time permits, discuss the foll., otherwise call for a separate meeting later this week to discuss:
(a) 1:1 mapping of NSI:NSSI
(b) NSSI information available in ONAP in case of external NSSMF, NSSMF APIs
(c) Implicaton of (3) in SO (1 top-most NSSMF?)
Status of Open Action Items
| S.No. | Action | Responsible | Status |
|---|---|---|---|
| 20191129_1 | Check for any differences in the text in 3GPP TS 28.530, 531 and 541 latest versions & the earlier ones | Swami & Borislav | Closed, details shared in updated doc sent on Dec 15th |
| 20191129_2 | Circulate the document with the various comments from ArchCom & community, & points for discussions, along with the scope definition | Swami, all to provide inputs/feedback | Ongoing, updated version circulated, further updates to be made based on ongoing discussions |
| 20191202_1 | Provide a couple of practical examples of CST/communication service request contents, NST contents and where more than 1 NST can fulfill the needs of a communication service request. | Chuyi Guo | Open, template to be shared by Swami |
| 20191202_2 | Based on above input from Chuyi Guo, Borislav to provide feedback on feasibility using existing service reference and nested services approach | Borislav | Closed, for Frankfurt, we will go with allotted resources approach, however, we will take up the discussion in Jan for the longer-term approach |
| 20191209_1 | Send an invite for the modeling discussion on Dec 11th | Swami | Closed |
| 20191209_2 | Send an invite for EXT-API impact discussion | Swami | Closed, EXT-API discussed with Adrian |
| 20191211_1 | Check for possible alternatives to avoid use of E2EST | Borislav, Swami | Closed, we will go with E2EST, possibility of changing the name to be considered |
| 20191211_2 | Respond to Borislav's observations and concerns on Allotted Resources approach | Chuyi Guo, Chen Chuanyu | Closed for Frankfurt, for Guilin and beyond, discussion to be taken up from Jan |
| 20191211_3 | Propose possible alternative to Allotted Resources approach | Borislav | Closed for Frankfurt, for Guilin and beyond, discussion to be taken up from Jan |
| 20191211_4 | Collect relevant information on E2EST, Allotted Resources and possible alternatives - pros and cons to enable conclusion on Dec 16th | Swami (with inputs from Borislav, Chuyi Guo, Chen Chuanyu and Seshu) | Closed, details collected and discussed on Dec 16th |
Notes and Actions
- The topic of the use of E2EST was discussed along with the strategy to use service reference vs allotted resources.
- Considering the impacts and the timelines, it was decided to go with the allotted resources approach for Frankfurt release. Due to the current limitation in ONAP SO w.r.to handling a service creation request, E2EST (or a renamed version) shall be used for Frankfurt release.
- For the longer term, discussions on the best solution would be started in January, considering the (enhanced) service reference, as well as the service resolver proposed by Orange. Of course, this would also have impacts in SO.
- It was recommended that 1:1 mapping between NSI and NSSI to be considered, even though the 1:1 mapping between NSMF and NSSMF may be realized using some simplifying assumptions for Frankfurt. This would also depend on the information exposed by the external NSSMF to ONAP. So, this aspect has to be clarified for Frankfurt. With respect to the longer term interoperability of ONAP with any external NSSMF, it will be discussed subsequently (e.g., discovery of NSSMF capabilities and ONAP adapting accordingly, etc.).
- In order that the OOF performs a meaningful selection of NSI as well as suggestion on NSSI, ONAP should have sufficient information from the external NSSMF. So this info has been requested.
- Since we are going with the existing capabilities of ONAP from a modeling perspective, we will present the details about our use case to the Modeling Sub-committee in Jan.
- No consensus was reached reg. S-NSSAI generation, for now, it is assumed that the CSMF will generate it.
We will have the weekly call on 23rd Dec, but the call on 30th Dec will be cancelled.
| S.No. | Action | Responsible |
|---|---|---|
| 20191216_1 | Make necessary updates to the modeling (e.g., rename E2EST if you have a better suggestion), and add more illustration/explanation on:
| Chuyi Guo, Chen Chuanyu |
| 20191216_2 | Share details of the info about the NSSIs and its constituents that is exposed by the external NSSMF with ONAP, share details of the APIs | Chen Chuanyu |
| 20191216_3 | Finalize the schema changes in A&AI, share it within the team for review | Chen Chuanyu |
| 20191216_4 | Respond to the queries on interaction between EXT-API and UUI | Lin Meng |
| 20191216_5 | Finalize the NSI:NSSI mapping for Frankfurt | Swami, Chuyi Guo, Chen Chuanyu, Seshu |