Re: [Teas] Network Slicing design team definitions- transport slice

["BRUNGARD, DEBORAH A" <[e-mail]>]

Hi,
(individual)

I’m still very confused by introducing the new term “transport” for IETF work on TE. As Young noted, it will be confused with the archaic term transport for many and it’s a misuse of 3GPP terms if it is not limited to 3GPP’s transport. Please look at the discussion thread on ACTN.

Ccamp and RAW cover wireless technologies so TEAS work is not limited to 3GPP “transport”. This work should be generic - a base - for other IETF work.

Before defining yet another term, why is the term Network Slice or TE network slice not acceptable? TE network slice will distinguish this work is for TE domains (TEAS charter work), network slice will align with IETF’s terms used in other groups. I don’t think TE transport slice is acceptable either - when we already have existing terminology that is concise.

I think you are confusing with the technology - especially if you are introducing new terms for all parts of the network - RAN slices, Core slices, Transport slice - TEAS work has never been constrained by the technology domain. “Network” is the network, no matter the division. There’s no need for all these terms.

Deborah
(Individual)

Sent from my iPhone

On Sep 1, 2020, at 2:17 PM, Rokui, Reza (Nokia - CA/Ottawa) <[e-mail]> wrote:


All,

Please note that 3GPP does not use the term “RAN Slice” or “Core slice”. Instead they use term “Slice Subnet”.
Please refer to the following 3GPP documents for example for more info:


  *   TS 28.531
  *   TR 28.801

Also see the following (taken from http://www.3gpp.org/NEWS-EVENTS/3GPP-NEWS/1951-SA5_5G<https://urldefense.proofpoint.com/v2/url?u=http-3A__www.3gpp.org_NEWS-2DEVENTS_3GPP-2DNEWS_1951-2DSA5-5F5G&d=DwMGaQ&c=LFYZ-o9_HUMeMTSQicvjIg&r=6UhGpW9lwi9dM7jYlxXD8w&m=9ijmMtGVBggFatq3IfGWDVh1iMG4UJB0uI4dafgi_Aw&s=1SLoaA3fRyNItAXx5nrYuHbWzRXhSUL_n5JpSvcB3sI&e=>).
The term NSI is for Network Slice instance and term NSSI is for Network Slice Subnet instance. As the picture shows, 3GPP does not differentiate between RAN and Core, all are called NSSI
The draft’s co-authors did not use the term “subnet” because at IETF this term (i.e. IP Subnet) has very clear meaning and using it in context of network slicing is confusing.

Also since the work at IETF is not just for 5G, we use the term “Transport slice”. If the use case is 5G, we also used RAN and Core Slices instead of NSSI to remove any confusion.


<image001.png>


From: Teas <[e-mail]> on behalf of Young Lee <[e-mail]>
Date: Tuesday, September 1, 2020 at 10:56 AM
To: "BRUNGARD, DEBORAH A" <[e-mail]>
Cc: TEAS WG <[e-mail]>
Subject: Re: [Teas] Network Slicing design team definitions- transport slice

Hi Deborah,

I also am wondering where ' transport slicing' came from. 3GPP uses network slicing in the following three subnets: RAN, CN, and TN; and accordingly it uses RAN slicng, CN slicing  and TN slicing.

There is always prefix "Network" is associated with slicing whether it is RAN, CN, or TN.

As you pointed out, when RFC 8453 was developed in TEAS WG, we originally used the term Transport Network Slicing, but later this term was replaced with Traffic Engineered (TE) Network slicing.

Transport slicing is a bit orphanish. It should be either TE network slicing or Transport network slicing. In TEAS WG, TE network slicing might be more consistent with RFC 8453. Just for your reference, the following terminology has been defined in RFC 8453:


TE Network Slicing:  In the context of ACTN, a TE network slice is a

      collection of resources that is used to establish a logically

      dedicated virtual network over one or more TE networks.  TE

      network slicing allows a network operator to provide dedicated

      virtual networks for applications/customers over a common network

      infrastructure.  The logically dedicated resources are a part of

      the larger common network infrastructures that are shared among

      various TE network slice instances, which are the end-to-end

      realization of TE network slicing, consisting of the combination

      of physically or logically dedicated resources.

Thanks.
Young



2020년 9월 1일 (화) 오후 11:31, BRUNGARD, DEBORAH A <[e-mail]<mailto:[e-mail]>>님이 작성:
Hi,
(speaking as an individual)

Picking up on Jari’s comment “sometimes people want to use labels that are a bit more imprecise, but I think we should strive to avoid such labels as much as we can”, can you give more background on the choice of the term “transport slice”?

I noted while you have text on the use of “transport” by 3GPP as specifying a very specific sub-network in the radio network (which excludes the handoff to the rest of the network), you referenced RFC5921 (MPLS-TP) for the term “transport”.

There was a long thread of mail on ACTN, RFC8453, to move away from the term “transport” to “TE networks” which was agreed to be more specific/appropriate for TEAS. Considering IETF is already using the term “network slicing” (including TEAS RFC8453) and 3GPP use of the term, I don’t understand the use of “transport” in this document.

Considering IETF already uses the term “network slice” and 3GPP uses the term (and all the industry SDOs), I am concerned we will be confusing the industry, especially as “transport” is an imprecise term for TEAS work. Is this work a subset and scoped only to the 3GPP “transport” cloud (and implied use of MPLS-TP)? Or is it planned to be generic (for all TE networks)?

Deborah
(speaking as individual)
_______________________________________________
Teas mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/teas<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ietf.org_mailman_listinfo_teas&d=DwMGaQ&c=LFYZ-o9_HUMeMTSQicvjIg&r=6UhGpW9lwi9dM7jYlxXD8w&m=9ijmMtGVBggFatq3IfGWDVh1iMG4UJB0uI4dafgi_Aw&s=LvYkDPlAn-K0sjB9D2KX_ocsYQqekQuDux4QILJryE0&e=>