[Teas] networking slicing design team results + call for feedback

Jari Arkko <jari.arkko@piuha.net> Thu, 02 April 2020 09:37 UTC

From: Jari Arkko <jari.arkko@piuha.net>
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Mac OS X Mail 10.3 \(3273\))
Message-Id: <FF568B3C-DC75-4E5F-A393-D2CB3F83AF97@piuha.net>
Date: Thu, 02 Apr 2020 12:36:56 +0300
To: teas@ietf.org
Archived-At: <https://mailarchive.ietf.org/arch/msg/teas/CCnfzDAgk_P7Z26Mr2WV5ezcN4g>
Subject: [Teas] networking slicing design team results + call for feedback
Precedence: list

Hi,

As you know, we’ve had a design team looking at how IETF VPN and TE technologies can be used to assist network slicing; to provide a framework [1]. The design team has been running for a couple of months, enough to have some first, very early results. We have two documents, one about definitions and another one about a framework of how slicing, underlying technologies, controllers, etc. fit together. The central concept is that one of a “transport slice”, which we believe would be useful as a component in various people building their networks. A transport slice is nothing magical, it is simply the service that many IETF technologies are already offering. And we can of course keep improving them.

It is important to understand that the design team is not here to develop new VPN/TE technologies, not here to focus on specific new characteristics that one may want from one’s VPNs, not here to define 5G slicing, not here to look at compute and services, and hopefully not here to boil the ocean. We’re also not done, as noted the documents that we have are early, and perhaps more importantly, there are still other things to develop. For instance, presumably one would like to have a data model or an interface with which one can request transport slices. Such an interface may partially exist already but the design team seems to believe that at least some enhancements will be needed. That’s all future work, either in the design team or as part of existing tech being improved. We also do not at the moment have a document that would point to all the relevant existing technologies that can be used to realise the framework or other frameworks across technologies for other reasons (e.g., enhanced-vpn that looked at enhanced isolation and also has a very good documentation of the existing IETF VPN technologies). Those are all forthcoming work as well.

At this point we’d love to get feedback on these documents though:

https://tools.ietf.org/html/draft-nsdt-teas-transport-slice-definition-01
https://tools.ietf.org/html/draft-nsdt-teas-ns-framework-02

Jari

[1] https://mailarchive.ietf.org/arch/msg/teas/jiHWXU_i5kK5BzjRffFbYnbZJfs/

[Teas] networking slicing design team results + c… Jari Arkko