Re: [Teas] Decision point on scope of draft-ietf-teas-ietf-network-slices

[Aihua Guo <[e-mail]>]

Hi Igor, all,

To say network slicing is only for 5G or traffic is only IP, is a bit too
narrow in scope, IMO.

- There are use cases for non-IP network, such as OTN, to carry both IP and
non-IP traffic from end user to the cloud or between DC and DC.
- There are use cases for non-IP traffic, such as 4k/8k video to be carried
end to end on circuit-emulated IP networks or circuit-native networks such
as OTN and DWDM.
- There are even emerging use cases for optical satellite communications
where the network links between LEO/GEO satellites, space station and the
ground are all optical. Such networks can carry IP or non-IP data. This
scenario is actively being discussed in this week’s OFC conference.

In all the above use cases, it can be envisioned that there is a need for a
“consumer” who needs services from operators of these networks, to ask for
some kind of service as a group of network resources with specific level of
quality.

Imagine if I am someone who owns an OTN network or an optical satellite
network, and I provide slices as services to which consumer who needs them
and is willing to pay.

We understand that networking of IP, optical, etc. are all in the scope for
IETF and TEAS. If the framework for network slicing is sufficiently
generic, as it should, then why we cannot apply the same terminology
towards those use cases, and construct it as a “slice”? Why we need to self
restrict the framework to only 5G or IP if we can make it more generic?

Yours,
Aihua

On Mon, Mar 7, 2022 at 10:19 AM Igor Bryskin <i_bryskin=
[e-mail]> wrote:

> Daniele,
>
> We all agreed that OTN  does not bring anything new into 5G that, for
> example,  was not happening  in 4G. OTN slicing work is motivated in some
> way, at least, because NS framework stipulates that the framework  is not
> limited to 5G. In fact, it seems to be not limited to anything (application
> -wise or layering -wise) to a point that it is not really clear what could
> not be considered a network slice.
>
> However,  If the framework was focused on 5G (ideallt going even further
> and stating  that said 5G style network slicing happenning in IP layer
> only, which I believe is the case) ,  OTn slicing would have nothing to do
> with the framework (and the term  slicing) and would have to rely on its
> own merits ( which,  as you mentioned, you have doubts about yourself) and
> also  explain why it coincides in time with NS work triggered and inspired
> entirely by 5G.
>
> Cheers,
> Igor
>
> Sent from Yahoo Mail on Android
> <https://go.onelink.me/107872968?pid=InProduct&c=Global_Internal_YGrowth_AndroidEmailSig__AndroidUsers&af_wl=ym&af_sub1=Internal&af_sub2=Global_YGrowth&af_sub3=EmailSignature>
>
> On Mon, Mar 7, 2022 at 12:01 PM, Daniele Ceccarelli
> <daniele.ceccarelli=[e-mail]> wrote:
> _______________________________________________
> Teas mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/teas
>
> _______________________________________________
> Teas mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/teas
>