Agenda:
- PNFD - the onboarded model (including non-MANO artifacts) and the internal model
Michela Bevilacqua and Zu Qiang (Ericsson) presented a proposal for the onboarding PNF packaging format that includes an extended set of artifacts. The proposal is backed by Ericsson, Nokia and Amdocs. Besides the detailed description of the onboarding format, the proposal also suggests a concrete way to map the onboarding model to the ONAP internally distributed service CSAR.
Anatoly Katzman: The onboarding part of the proposal is fine, as long as it is consistent, well-documented, and agreed across the ETSI community. However the "internal ONAP" part seems to be over-engineered, it introduces numerous new artifact types long before these artifacts are actually provided and probably before the ONAP components are really ready to consume them. For the purposes of the Dublin release, ONAP would better start with a more basic model, and then extend it in subsequent releases.
Michela Bevilacqua , Zu Qiang (Ericsson) : the suggested artifact types are mature, totally ONAP-compliant, and will be surely consumed by ONAP components (DCAE and CDN-C) during Dublin. It is safe to define these types now.
TODO: reach out to the DCAE and SDN-C teams, get their input on this. - ETSI NSD vs ONAP service descriptor by Thinh Nguyenphu
We didn't have enough time to cover this topic, will do it on the next DM call - For offline review and discussion (as suggested by Thinh Nguyenphu) - alignment with ETSI SOL001.
In the Dublin release, ONAP is going to align its onboarding data model and part of the internal data model with ETSI SOL001 2.5.1. Thinh Nguyenphu kindly provided an overview of the relevant type definitions on the pages ONAP R4+ Onboarding NSD based on ETSI NFV SOL001 v2.5.1, ONAP R4+ Onboarding VNFD based on ETSI NFV SOL001, and ONAP R4+ Onboarding PNFD based on ETSI NFV SOL001. The community members are all invited to review the pages and to share their thoughts in comments there.