Re: [Yot] draft-veillette-core-yang-library update

[Juergen Schoenwaelder <[e-mail]>]

The main NMDA document is draft-ietf-netmod-revised-datastores-10.txt,
meanwhile in the RFC editor queue. It may be a good idea to follow
NETMOD and NETCONF if people work on new protocols. ;-)

/js

On Tue, Jan 23, 2018 at 04:51:35PM +0000, Michel Veillette wrote:
> Thanks for this warning Juergen
> 
> I was aware of the NMDA acronym but not about the specific impacts of this approach on protocols and on the yang library.
> 
> For now, I will just add a comment to the draft to request an alignment with NMDA and draft-ietf-netconf-rfc7895bis-03. CoMI also need to be reviewed for NMDA support. I assume that draft-dsdt-nmda-guidelines-01, draft-ietf-netconf-rfc7895bis-03 and draft-ietf-netconf-nmda-restconf-02 are the main documents to review before proceeding?
> 
> Regards,
> Michel
> 
> -----Original Message-----
> From: Juergen Schoenwaelder [mailto:[e-mail]] 
> Sent: Monday, January 22, 2018 5:25 PM
> To: Michel Veillette <[e-mail]>
> Cc: [e-mail]
> Subject: Re: [Yot] draft-veillette-core-yang-library update
> 
> I assume you are aware of draft-ietf-netconf-rfc7895bis-03.txt and the discussions ongoing in NETCONF. The idea is to obsolete the old YANG library defined in RFC 7895 and to replace it with a new YANG library that supports NMDA.
> 
> /js
> 
> On Mon, Jan 22, 2018 at 08:38:29PM +0000, Michel Veillette wrote:
> > Draft https://datatracker.ietf.org/doc/draft-veillette-core-yang-library/ will expire in two days.
> > If you have changes to propose to this YANG module, please send them so I can update the draft accordingly.
> > 
> > On the Trilliant side, we propose to add support for a hash as cashing mechanism.
> > The resulting 'module-set-id' will support the following datatypes.
> > - uint32, local library content identifier
> > - identityref, global library content identifier (i.e. SID)
> > - byte string, global library content identifier (e.g. SHA2)
> > 
> > Do we need to support all these approaches?
> > 
> >   *   The local identifier is easy to manage, compact (~ 1, 2, 3 bytes) but the client need to manage device specific caches
> >   *   The SID is more complex to manage (i.g. required an interaction with the SID registry), is relatively compact (~ 5 bytes) and allows the implementation of a single cache for all devices.
> >   *   The hash is easy to manage but not as compact (~20 bytes)
> > 
> > 
> > [cid:[e-mail]0BB7E0]
> > 
> > Michel Veillette
> > System Architecture Director
> > Trilliant Inc.
> > Tel: [phone]
> > https://trilliantinc.com/
> > 
> > 
> > 
> 
> 
> 
> > _______________________________________________
> > Yot mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/yot
> 
> 
> -- 
> Juergen Schoenwaelder           Jacobs University Bremen gGmbH
> Phone: [phone]         Campus Ring 1 | 28759 Bremen | Germany
> Fax:   [phone]         <https://www.jacobs-university.de/>

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: [phone]         Campus Ring 1 | 28759 Bremen | Germany
Fax:   [phone]         <https://www.jacobs-university.de/>