search

samans / draft-ybam-rfc8561bis

Internet draft to modify rfc8561
Other
1 stars 4 forks source link

Hardware model #33

Closed samans closed 1 month ago

samans commented 9 months ago

How to leverage ietf-hardware for microwave nodes?

Are there microwave specific objects that need to be added to an augmentation of ietf-hardware?

Here is an example of how the BBF is augmenting ietf-hardware for transceivers.

https://github.com/BroadbandForum/yang/blob/master/standard/equipment/bbf-hardware-transceivers.yang

samans commented 9 months ago

Here is the complete set of YANG. Look at the directory .\yang-TR-383-v1.7.0.zip\yang-TR-383-v1.7.0\standard\equipment https://github.com/BroadbandForum/yang/archive/refs/tags/TR-383-v1.7.0.zip

DanielaSpreafico commented 7 months ago

I add some examples of hardware 'components' for a MW node. The assumption is to use the 'ietf-hardware' model as defined on the RFC 8348, augmenting it as specific technology attributes for mw. MW-Node-HWExample.pptx

As an alternative, discussed on last meeting (18th April), there is a model on progress on IETF IVY Working Group (see ietf-ivy-wg/network-inventory-yang (github.com) ). We need to take into account that this model is planned to be used for NBI interface on a Controller.

Discussion and comments are welcome.

DanielaSpreafico commented 6 months ago

I add a JSON as example for the previous proposal: mw-hw.json

italobusi commented 6 months ago

One doubt on my side: do we need to represent in the hardware inventory ODU/IDU chassis type or just use the chassis component class for both? Is there something specific which is needed to differentiated ODU and IDU chassis, other than they location?

From the figure, it seems that the relationship between IDU and ODUs is similar to the relationship between a main shelf and the sub-shelves in optical NEs.

Regarding the modelling of the ports and of the intra-NE cables, there are few open issues in IVY WG:

I think it would be good if we keep alignment with the on-going work in IVY WG

jrebiffe commented 5 months ago

This morning during Microwave meetings series, @italobusi, @DanielaSpreafico, Julien, Danilo and I, we said that starting a new draft would be a way to go. I took that point to initiate it.

DanielaSpreafico commented 4 months ago

Following a discussion on the yesterday meeting (2024-07-04) I upload an update to summarize the topics to agree.

MW-Node-HWExample-It2.pptx

Main points:

  1. how to model the ODU :

    • using 'chassis' class ? Not allowed on a 'container'
    • using a 'module' class ? This is a proposal from Danilo
    • introducing a new IDENTITY 'mw-odu' ?

  2. how to model the connection of an ODU to a port on IDU? This both in case the connection is using an ethernet port or an IF cable:

    • defining the 'container' (where the ODU could be contained) as a child of the port component connected to?
    • adding something new?

  3. how to add (or change) a new item HW by NETCONF client?

  4. how about the use of the ethernet port connecte or not connected to an ODU for a service (Bridge, etc)?

JonasAhl commented 4 months ago

I've added another example how ODUs can be modelled on page 7 in the attached presentation. MW-Node-HWExample-v3.pptx

italobusi commented 2 months ago

See also: https://github.com/ietf-ivy-wg/network-inventory-yang/issues/64

See also some previous discussions in https://github.com/ietf-ccamp-wg/ietf-network-inventory/issues/6 which has been closed as superseded by https://github.com/ietf-ivy-wg/network-inventory-yang/issues/64

samans commented 1 month ago

Continue the discussion in IVY.