Re: [Teas] Definitions of VTN and Slice Aggregate in the drafts

"Dongjie (Jimmy)" <jie.dong@huawei.com> Tue, 23 March 2021 12:46 UTC

Return-Path: <jie.dong@huawei.com>
X-Original-To: teas@ietfa.amsl.com
Delivered-To: teas@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3D0C93A0930 for <teas@ietfa.amsl.com>; Tue, 23 Mar 2021 05:46:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level:
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8IOCJLC0FOMx for <teas@ietfa.amsl.com>; Tue, 23 Mar 2021 05:46:15 -0700 (PDT)
Received: from frasgout.his.huawei.com (frasgout.his.huawei.com [185.176.79.56]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8D6AE3A0934 for <teas@ietf.org>; Tue, 23 Mar 2021 05:46:14 -0700 (PDT)
Received: from fraeml713-chm.china.huawei.com (unknown [172.18.147.207]) by frasgout.his.huawei.com (SkyGuard) with ESMTP id 4F4WFg19Ghz681wB; Tue, 23 Mar 2021 20:41:27 +0800 (CST)
Received: from dggeme754-chm.china.huawei.com (10.3.19.100) by fraeml713-chm.china.huawei.com (10.206.15.32) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256) id 15.1.2106.2; Tue, 23 Mar 2021 13:46:09 +0100
Received: from dggeme754-chm.china.huawei.com ([10.6.80.77]) by dggeme754-chm.china.huawei.com ([10.6.80.77]) with mapi id 15.01.2106.013; Tue, 23 Mar 2021 20:46:07 +0800
From: "Dongjie (Jimmy)" <jie.dong@huawei.com>
To: Daniele Ceccarelli <daniele.ceccarelli@ericsson.com>, "Belotti, Sergio (Nokia - IT/Vimercate)" <sergio.belotti@nokia.com>, Young Lee <younglee.tx@gmail.com>
CC: "adrian@olddog.co.uk" <adrian@olddog.co.uk>, TEAS WG <teas@ietf.org>, Italo Busi <Italo.Busi@huawei.com>
Thread-Topic: [Teas] Definitions of VTN and Slice Aggregate in the drafts
Thread-Index: AdcVDiGdwCInSefCSJeE9AkCi8oHbQAfwOJAADGwKWD//8qsgIAGOTyAgADY0wCAAeKWAP//WjsQgADY4oD//bwmEIAFnREA//k2KBA=
Date: Tue, 23 Mar 2021 12:46:07 +0000
Message-ID: <9e955da5165e433686772dfff7ad6d2f@huawei.com>
References: <f1ca882deea4405981067c6d16a46d19@huawei.com> <AM0PR07MB549027D23F3542C55306E3E391919@AM0PR07MB5490.eurprd07.prod.outlook.com> <b3803b28b4064a78bc78c7b727dd937e@huawei.com> <0c5801d7167c$4b700fb0$e2502f10$@olddog.co.uk> <AM0PR07MB5490BB193E704133CF1E7776916C9@AM0PR07MB5490.eurprd07.prod.outlook.com> <CAGHSPWNb2uEGXkL8Os=eXscCz8UMVj6Lt9YzmCGtEkx06Yx9CQ@mail.gmail.com> <HE1PR0701MB2282B7E8453FB12E7DF5999CF06A9@HE1PR0701MB2282.eurprd07.prod.outlook.com> <23854013cac1407da72cc5efd9eda75b@huawei.com> <AM0PR07MB5490EE42841D967675132F63916A9@AM0PR07MB5490.eurprd07.prod.outlook.com> <d1c696cb520a41c0af86186cf03d1f1d@huawei.com> <HE1PR0701MB2282411457473E0A5C91DDE7F0689@HE1PR0701MB2282.eurprd07.prod.outlook.com>
In-Reply-To: <HE1PR0701MB2282411457473E0A5C91DDE7F0689@HE1PR0701MB2282.eurprd07.prod.outlook.com>
Accept-Language: en-US, zh-CN
Content-Language: zh-CN
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
x-originating-ip: [10.108.243.143]
Content-Type: multipart/alternative; boundary="_000_9e955da5165e433686772dfff7ad6d2fhuaweicom_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <https://mailarchive.ietf.org/arch/msg/teas/LvBeviFVdBmxGtYKQUcVkPr9fHk>
Subject: Re: [Teas] Definitions of VTN and Slice Aggregate in the drafts
X-BeenThere: teas@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: Traffic Engineering Architecture and Signaling working group discussion list <teas.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/teas>, <mailto:teas-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/teas/>
List-Post: <mailto:teas@ietf.org>
List-Help: <mailto:teas-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/teas>, <mailto:teas-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 23 Mar 2021 12:46:20 -0000

Hi Daniele,

From your statement, I see that currently VN can be instantiated using TE tunnels, the augmentation with other mechanism is possible and may be discussed further in the WG.

This confirms my understanding that VN as a view on the network controller is decoupled from its realization, the instantiation of a VN in the network is be based on specific technologies, e.g. TE tunnels, etc.

To help understanding, you may consider VTN as one instantiation of VN in the network.

When a VTN is created, the topology and the set of resources are determined, and the states of the VTN topology and resources are maintained on the network nodes involved.

BTW, could you also provide thoughts about other similar terms: slice aggregate and slice policy?

Best regards,
Jie

From: Daniele Ceccarelli [mailto:daniele.ceccarelli@ericsson.com]
Sent: Friday, March 19, 2021 8:39 PM
To: Dongjie (Jimmy) <jie.dong@huawei.com>; Belotti, Sergio (Nokia - IT/Vimercate) <sergio.belotti@nokia.com>; Young Lee <younglee.tx@gmail.com>
Cc: adrian@olddog.co.uk; TEAS WG <teas@ietf.org>; Italo Busi <Italo.Busi@huawei.com>
Subject: RE: [Teas] Definitions of VTN and Slice Aggregate in the drafts

Hi Jie,

This is the type of augmentation I was referring to. It could be worth considering adding non TE scenarios to ACTN VNs.
VNs could be composed by TE tunnels or non TE constructs/policies.
In the end if a VN member is composed by a TE tunnel, a flex ago o a routing policy doesn’t really make a difference from the VN point of view, simply the path between the end points will not be predetermined and resources won’t be reserved.

BR
Daniele

From: Dongjie (Jimmy) <jie.dong@huawei.com<mailto:jie.dong@huawei.com>>
Sent: den 18 mars 2021 17:21
To: Belotti, Sergio (Nokia - IT/Vimercate) <sergio.belotti@nokia.com<mailto:sergio.belotti@nokia.com>>; Daniele Ceccarelli <daniele.ceccarelli@ericsson.com<mailto:daniele.ceccarelli@ericsson.com>>; Young Lee <younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>>
Cc: adrian@olddog.co.uk<mailto:adrian@olddog.co.uk>; TEAS WG <teas@ietf.org<mailto:teas@ietf.org>>; Italo Busi <Italo.Busi@huawei.com<mailto:Italo.Busi@huawei.com>>
Subject: RE: [Teas] Definitions of VTN and Slice Aggregate in the drafts

Hi Sergio,

I think we may look at different components/layers of the network: you are focusing on the VN model and the related models, while I’m talking about the network states instantiated in the network nodes.

The network view provided by the VN model resides on the network controllers (e.g. PNC, MDSC, CNC in the ACTN architecture). As you mentioned the instantiation of the VN model on network nodes can be based on existing tools (the TE tunnels).

VTN is about the states in the network nodes. Each involved network node allocates and maintains VTN specific topology and/or resource state, which are used to perform VTN-specific packet forwarding. This could be done based on MT or Flex-Algo or similar control plane mechanisms, with SR or MPLS or IPv6 as the data plane, although TE tunnels may also be one option.

Best regards,
Jie

From: Belotti, Sergio (Nokia - IT/Vimercate) [mailto:sergio.belotti@nokia.com]
Sent: Wednesday, March 17, 2021 5:30 PM
To: Dongjie (Jimmy) <jie.dong@huawei.com<mailto:jie.dong@huawei.com>>; Daniele Ceccarelli <daniele.ceccarelli@ericsson.com<mailto:daniele.ceccarelli@ericsson.com>>; Young Lee <younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>>
Cc: adrian@olddog.co.uk<mailto:adrian@olddog.co.uk>; TEAS WG <teas@ietf.org<mailto:teas@ietf.org>>; Italo Busi <Italo.Busi@huawei.com<mailto:Italo.Busi@huawei.com>>
Subject: RE: [Teas] Definitions of VTN and Slice Aggregate in the drafts

Hi Jie,
I do not think Adrian said exactly what you said: he agreed on the fact that the concept is used in different context. This is not in discussion.
BTW the abstraction that you mention is depending of the level customer want to apply, nothing prevent customer to ask for a native topology that nothing is if not the underlay topology that you mention. Section 5 of RFC 8453 explains very the concept.
About what is used to create the required topology is based in what chairs suggested to do, so use the tool already there in TEAS. What do you mean with “network resources” and what would be the model you’d like to use to create your VTN?
“Thus in my understanding the VN/VN model and the VTN are used in different layers of the network for different purposes”
Not clear what you mean here frankly speaking , the underlay or native topology is in the layer that exist depending on the technology is used , there is no difference here based on layers.
Your last concern about the “generality” of the term, is not a valid concern: virtual, abstract, are terms that many time are used in different way, but what I mention as reference is clearly defined, and used to create a specific YANG model, not just words, it is not generic .
Maybe an alignment with your colleagues that worked hard on VN could help the common understanding.

Thanks
Regards, Sergio



From: Dongjie (Jimmy) <jie.dong@huawei.com<mailto:jie.dong@huawei.com>>
Sent: Wednesday, March 17, 2021 10:07 AM
To: Daniele Ceccarelli <daniele.ceccarelli@ericsson.com<mailto:daniele.ceccarelli@ericsson.com>>; Young Lee <younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>>; Belotti, Sergio (Nokia - IT/Vimercate) <sergio.belotti@nokia.com<mailto:sergio.belotti@nokia.com>>
Cc: adrian@olddog.co.uk<mailto:adrian@olddog.co.uk>; TEAS WG <teas@ietf.org<mailto:teas@ietf.org>>
Subject: RE: [Teas] Definitions of VTN and Slice Aggregate in the drafts

Hi,

As Adrian and I explained in previous mail, VN is an abstracted view of the network provided to a customer, this can be found in the definition of VN in RFC 8453:

  Virtual Network (VN):  A VN is a network provided by a service provider to a customer for the customer to use in any way it wants as though it was a physical network.  There are two views of a VN as follows:

    The VN can be abstracted as a set of edge-to-edge links (a Type 1 VN).

    The VN can also be abstracted as a topology of virtual nodes and virtual links (a Type 2 VN)

Thus the key of VN is “abstraction”. IMO this is different from VTN, which refers to the realization of a virtual underlay network based on operator’s physical network, no abstraction is used in the creation of VTN.

     “Virtual Transport Network (VTN) which has a customized network topology and a set of dedicated or shared resources in the underlay network.

Moreover, in draft-ietf-teas-actn-vn-yang, the instantiation of VN is based on the TE topology model and the TE tunnel model. While VTN is created as a subset of the underlay network topology and network resources, its realization is not bound to the TE tunnels.

Thus in my understanding the VN/VN model and the VTN are used in different layers of the network for different purposes.

Also please note my concern in my previous mail: another consideration was that the term VN is generic (not only defined in RFC 8453), thus could mean different things in different context. Thus we choose to use a dedicated term for the virtual underlay network.

Best regards,
Jie

From: Daniele Ceccarelli [mailto:daniele.ceccarelli@ericsson.com]
Sent: Wednesday, March 17, 2021 2:28 PM
To: Young Lee <younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>>; Belotti, Sergio (Nokia - IT/Vimercate) <sergio.belotti@nokia.com<mailto:sergio.belotti@nokia.com>>
Cc: adrian@olddog.co.uk<mailto:adrian@olddog.co.uk>; Dongjie (Jimmy) <jie.dong@huawei.com<mailto:jie.dong@huawei.com>>; TEAS WG <teas@ietf.org<mailto:teas@ietf.org>>
Subject: RE: [Teas] Definitions of VTN and Slice Aggregate in the drafts

Hi,

A +1 on what Young and Sergio said.

The definition of VTN says:

   o  A VTN is a virtual underlay network that connects customer edge
      points.  The VTN has the capability to deliver the performance
      characteristics required by an enhanced VPN customer and to
      provide isolation between separate VPN+ instances.

The only thing that is missing in the VN today is the isolation part (which btw seems to be a quite controversial topic not yet fully supported by the WG). However it’s possible to create a VN at different layers using different technologies. If e.g. a VN is created using a specific ODU in an OTN network, that’s absolutely isolated from other traffic.

Could you please help us better understand what is missing in the VN to meet the VTN requirements? If anything is missing I would suggest to augment the VN model rather than creating a completely new one.
A while ago Lou and Pavan suggested to have the VN definition independent from ACTN so that it could have been used outside ACTN context…apparently it was a good move 😊

BR
Daniele


From: Young Lee <younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>>
Sent: den 16 mars 2021 02:40
To: Belotti, Sergio (Nokia - IT/Vimercate) <sergio.belotti@nokia.com<mailto:sergio.belotti@nokia.com>>
Cc: adrian@olddog.co.uk<mailto:adrian@olddog.co.uk>; Dongjie (Jimmy) <jie.dong@huawei.com<mailto:jie.dong@huawei.com>>; Daniele Ceccarelli <daniele.ceccarelli@ericsson.com<mailto:daniele.ceccarelli@ericsson.com>>; TEAS WG <teas@ietf.org<mailto:teas@ietf.org>>; Young Lee <younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>>
Subject: Re: [Teas] Definitions of VTN and Slice Aggregate in the drafts

Hi all,

I concur with Sergio in this thread. VN has different levels of granularity such that it can be at the same level as VTN. As Sergio well pointed out, VN type 2 is a type of Virtual Network "topology" from which operators can make use of it for its network operation such as mpls, SR, etc.

From a modeling standpoint, all yang models pertaining to VN (such as VN yang, TE service mapping, etc.) can be used for VTN contexts as is or can be augmented for VTN use from existing VN models if need be.

Thanks.
Young

2021년 3월 15일 (월) 오후 9:44, Belotti, Sergio (Nokia - IT/Vimercate) <sergio.belotti@nokia.com<mailto:sergio.belotti@nokia.com>>님이 작성:
Hi Adrian, Jie,

Thanks for your replies.
Sincerely , while I agree on the fact that VN and VTN are introduced for different purpose, nothing prevent to use the “same” concept for different context as soon as the concept is covering multiple requirement!
So question is again what in VN is preventing to be used in an enhanced VPN.
Adrian underlined that ACTN tool/mechanism is able to provide a network slice and then an enhanced VPN, and so I do not understand why a VN type 2 for example should be different to what Jie described as VTN.
Moreover , for VN, exists already a YANG to create a VN and in parallel a VN service mapping as well, so question is : is this model able to support a VTN as well or, even if for a similar concept , we need to have a separate YANG model ?
A proliferation of different models are not good in my view without real needs.

Thanks in advance

Sergio


From: Adrian Farrel <adrian@olddog.co.uk<mailto:adrian@olddog.co.uk>>
Sent: Thursday, March 11, 2021 2:42 PM
To: 'Dongjie (Jimmy)' <jie.dong@huawei.com<mailto:jie.dong@huawei.com>>; Belotti, Sergio (Nokia - IT/Vimercate) <sergio.belotti@nokia.com<mailto:sergio.belotti@nokia.com>>
Cc: younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>; 'Daniele Ceccarelli' <daniele.ceccarelli@ericsson.com<mailto:daniele.ceccarelli@ericsson.com>>; teas@ietf.org<mailto:teas@ietf.org>
Subject: RE: [Teas] Definitions of VTN and Slice Aggregate in the drafts

The line that stand out from this conversation for me is:

> As for the VN concept defined in RFC 8453, to me the concept of VN and VTN looks similar, while they are introduced for different purposes.

From my perspective, a VN (ACTN) is a topology carved out of the network and presented to the ACTN consumer, while a VTN (VPN+) is a topology carved out of the network for its own use in delivering enhanced VPNs. That looks pretty much what Jie said: same concept but for different purposes.

FWIW, in draft-king-teas-applicability-actn-slicing we saw ACTN as a tool/mechanism that could be used to realise a network slice or an enhanced VPN. That is, it can build the virtual topology that underpins a network slice.


I think a number of concepts are converging within TEAS. This should be cause for celebration, although it is not always easy to get full convergence.

What we found when writing draft-king-teas-applicability-actn-slicing was that the terminology was not consistent. That, sadly, led us to have to define some of our own terms. I fully expect that as time progresses we will see continued convergence on terminology that will cause a number of us to have to update our drafts.

Cheers,
Adrian

From: Teas <teas-bounces@ietf.org<mailto:teas-bounces@ietf.org>> On Behalf Of Dongjie (Jimmy)
Sent: 11 March 2021 09:57
To: Belotti, Sergio (Nokia - IT/Vimercate) <sergio.belotti@nokia.com<mailto:sergio.belotti@nokia.com>>
Cc: younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>; Daniele Ceccarelli <daniele.ceccarelli@ericsson.com<mailto:daniele.ceccarelli@ericsson.com>>; teas@ietf.org<mailto:teas@ietf.org>
Subject: Re: [Teas] Definitions of VTN and Slice Aggregate in the drafts

Hi Sergio,

Thanks for your support on the terminology clarification. It is important to have consistent understanding about the terminology before we could understand the relationship between different proposals.

As for the VN concept defined in RFC 8453, to me the concept of VN and VTN looks similar, while they are introduced for different purposes.

VN in RFC 8453 is to provide the customer with an abstracted view of the network based on YANG models. There are two modes to provide such view to customer: abstracted as a set of edge-to-edge links, or abstracted as a topology or virtual nodes and virtual links.

While the term VTN was introduced to refer to the virtual underlay network in the layered architecture as described in draft-ietf-teas-enhanced-vpn, it is used to organize a subset of the network nodes and links, and allocate a set of network resources on these nodes and links to build a customized underlay network. Operator can instantiate multiple VTNs to carry different types of services, or the traffic of different customers. The customers does not necessarily need to be provided with the information of VTN.

Another consideration was that the term VN is generic (not only defined in RFC 8453), thus could mean different things in different context. Thus we choose to use a dedicated term for the virtual underlay network.

Hope this helps.

Best regards,
Jie

From: Belotti, Sergio (Nokia - IT/Vimercate) [mailto:sergio.belotti@nokia.com]
Sent: Wednesday, March 10, 2021 5:18 PM
To: Dongjie (Jimmy) <jie.dong@huawei.com<mailto:jie.dong@huawei.com>>
Cc: teas@ietf.org<mailto:teas@ietf.org>; Daniele Ceccarelli <daniele.ceccarelli@ericsson.com<mailto:daniele.ceccarelli@ericsson.com>>; younglee.tx@gmail.com<mailto:younglee.tx@gmail.com>
Subject: RE: Definitions of VTN and Slice Aggregate in the drafts

Hi Jie,
Clarification on terminology is always a good effort to do  thanks for that.
On the other hand I’d like to understand , about “VTN”, what is the difference in your VTN concept with respect what already defined in RFC 8453 in section 2.1?
In that section there is a description also of the two types of VN , customer can manage.
A parallel effort with draft-ietf-teas-actn-vn-yang permits to “create” VN of the desired type.
So my question is : what in your “new” VTN is different with respect what already defined related to VN to force using another term ?

Thanks
Sergio





From: Teas <teas-bounces@ietf.org<mailto:teas-bounces@ietf.org>> On Behalf Of Dongjie (Jimmy)
Sent: Tuesday, March 9, 2021 7:38 PM
To: teas@ietf.org<mailto:teas@ietf.org>
Subject: [Teas] Definitions of VTN and Slice Aggregate in the drafts

Hi all,

In today’s meeting, some people raised comments about the confusions caused by the multiple terms. To help the understanding with the term “VTN” and “Slice Aggregate”, I reread the text about their definitions and descriptions in each draft.

Although the terminologies are described in different ways, IMO the essential information behind is the same: a logical network construct consists of the topology attribute and the network resources attributes.

Some effort to resolve the terminology overlap could help to reduce the confusions and provide a consistent view on the framework and technologies for network slice realization.


Here are the text about the terms quoted from each draft:

- VTN (in draft-ietf-teas-enhanced-vpn-07):

VPN+ is built from a VPN overlay and an underlying Virtual Transport Network (VTN) which has a customized network topology and a set of dedicated or shared resources in the underlay network.

A VTN is a virtual underlay network that connects customer edge points. The VTN has the capability to deliver the performance characteristics required by an enhanced VPN customer and to provide isolation between separate VPN+ instances.


- Slice Aggregate (in draft-bestbar-teas-ns-packet-02)

This document introduces the notion of a slice aggregate which comprises of one of more IETF network slice traffic streams. It describes how a slice policy can be used to realize a slice aggregate by instantiating specific control and data plane behaviors on select topological elements in IP/MPLS networks.

When logical networks representing slice aggregates are realized on top of a shared physical network infrastructure, it is important to steer traffic on the specific network resources allocated for the slice aggregate.

Slice aggregate: a collection of packets that match a slice policy selection criteria and are given the same forwarding treatment; a slice aggregate comprises of one or more IETF network slice traffic streams; the mapping of one or more IETF network slices to a slice aggregate is maintained by the IETF Network Slice Controller.


- Slice policy (in draft-bestbar-teas-ns-packet-02, to help the understanding of Slice Aggregate)

Slice policy: a policy construct that enables instantiation of mechanisms in support of IETF network slice specific control and data plane behaviors on select topological elements; the enforcement of a slice policy results in the creation of a slice aggregate.

A slice policy topology may include all or a sub-set of the physical nodes and links of an IP/MPLS network; it may be comprised of dedicated and/or shared network resources (e.g., in terms of processing power, storage, and bandwidth).

Best regards,
Jie