Re: [altoext] [alto] i2aex BOF - SDN Use Case

Ken Gray <kgray@juniper.net> Sun, 25 March 2012 21:22 UTC

Return-Path: <kgray@juniper.net>
X-Original-To: altoext@ietfa.amsl.com
Delivered-To: altoext@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3759221F8455; Sun, 25 Mar 2012 14:22:50 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.599
X-Spam-Level:
X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bDQdMp+YxFKf; Sun, 25 Mar 2012 14:22:49 -0700 (PDT)
Received: from exprod7og124.obsmtp.com (exprod7og124.obsmtp.com [64.18.2.26]) by ietfa.amsl.com (Postfix) with ESMTP id 346BA21F8454; Sun, 25 Mar 2012 14:22:47 -0700 (PDT)
Received: from P-EMHUB02-HQ.jnpr.net ([66.129.224.36]) (using TLSv1) by exprod7ob124.postini.com ([64.18.6.12]) with SMTP ID DSNKT2+MpdkDW6fxohXlHztvDYtLitn2e4C+@postini.com; Sun, 25 Mar 2012 14:22:49 PDT
Received: from EMBX02-HQ.jnpr.net ([fe80::18fe:d666:b43e:f97e]) by P-EMHUB02-HQ.jnpr.net ([fe80::88f9:77fd:dfc:4d51%11]) with mapi; Sun, 25 Mar 2012 14:22:39 -0700
From: Ken Gray <kgray@juniper.net>
To: Ping Pan <ping@pingpan.org>, ZongNing <zongning@huawei.com>
Date: Sun, 25 Mar 2012 14:22:38 -0700
Thread-Topic: [altoext] [alto] i2aex BOF - SDN Use Case
Thread-Index: Ac0KzWeY0wcw/7QCRheoVMGLMylQsw==
Message-ID: <CB95020B.B3FD8%kgray@juniper.net>
In-Reply-To: <CAHEV9L3KjM6Fua6T0dGHRbQB91V3R1NS1jKuR2tawjcMf38=uQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
user-agent: Microsoft-MacOutlook/14.14.0.111121
acceptlanguage: en-US
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "altoext@ietf.org" <altoext@ietf.org>, "Vijay K. Gurbani" <vkg@bell-labs.com>, "alto@ietf.org" <alto@ietf.org>, Spencer Dawkins <spencer@wonderhamster.org>, Tina TSOU <Tina.Tsou.Zouting@huawei.com>
Subject: Re: [altoext] [alto] i2aex BOF - SDN Use Case
X-BeenThere: altoext@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Non-WG list for discussions related to ALTO Protocol Extensions <altoext.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/altoext>, <mailto:altoext-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/altoext>
List-Post: <mailto:altoext@ietf.org>
List-Help: <mailto:altoext-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/altoext>, <mailto:altoext-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 25 Mar 2012 21:22:50 -0000

I always thought the relationship of ALTO to SDN was to be one of perhaps multiple topological information sources.  In short, I don't see ALTO on top of SDN, I see it as a plug-in module under the category of topology.

/jumpingontosoapbox

SDN decisions aren't made in a vacuum 8^), so any SDN-like broker needs information on how to get from here (client or client by proxy of some querying service) to there (service).  The extensions to ALTO that make it preferential as such a tool are those that make it more dynamic (partial updates, push) and the separation of cost from something harder to manipulate (or that you don't want to manipulate, like IGP metric) …and maybe even express something like … policy.

Where I see there being a pending collision is when a tool like ALTO expresses multiple sub-services of an overall SDN solution (like topology and policy) …and how we account for that in SDNP.  Because the challenge of SDNP is to see ALTO, OF, PCEP, BGP-TE, overlays (NOV3 for example) …as tools (some of them interchangeable) used to describe a service to some application through a northbound API.

What SDN needs is a standardized interface for representing topology amongst the many candidates (there's at least one other source I can think of off the top of my head … BGP-TE …and of course, static maps) that allows for varying richness in functionality.

/jumpingoffsoapbox

From: Ping Pan <ping@pingpan.org<mailto:ping@pingpan.org>>
Date: Sun, 25 Mar 2012 05:04:37 -0700
To: ZongNing <zongning@huawei.com<mailto:zongning@huawei.com>>
Cc: "altoext@ietf.org<mailto:altoext@ietf.org>" <altoext@ietf.org<mailto:altoext@ietf.org>>, "Vijay K. Gurbani" <vkg@bell-labs.com<mailto:vkg@bell-labs.com>>, "alto@ietf.org<mailto:alto@ietf.org>" <alto@ietf.org<mailto:alto@ietf.org>>, Spencer Dawkins <spencer@wonderhamster.org<mailto:spencer@wonderhamster.org>>, Tina TSOU <Tina.Tsou.Zouting@huawei.com<mailto:Tina.Tsou.Zouting@huawei.com>>
Subject: Re: [altoext] [alto] i2aex BOF - SDN Use Case

Concur.

ALTO is currently used in support of P2P and CDN operation. It has the functionality to query the network topological data. However that's where the similarity with SDN ends. Running them together, how are they inter-related?

Also you have brought up the concept of SDN Domains. What are they? Who owns them? Why do we need them? For scaling? For admin? If for scaling, SDN controller intrinsically has no scaling problem per sa. For admin, SDN is to offer an overlay architecture on top of the network with a uniformed view from application perspective, which is opposite of having some sort of peering models. Please elaborate your thoughts.

On Sun, Mar 25, 2012 at 12:52 PM, ZongNing <zongning@huawei.com<mailto:zongning@huawei.com>> wrote:
Hi, Tina,

Very interesting topic! One comment after a quick review:
In page 5 in the slides, it seems that you prefer to put ALTO on top of SDN controller. I am wondering if the primary goal of ALTO-SDN is for better SDN control or better ALTO service?
If the goal is for better SDN, then how ALTO could help SDN? Receive information from SDN controller, then combine with other ALTO information and feedback to SDN?
Also in page 5, I don't understand why the data are forwarded from SDN to ALTO?
Would you please elaborate this? Thanks.

-Ning
________________________________________
From: altoext-bounces@ietf.org<mailto:altoext-bounces@ietf.org> [altoext-bounces@ietf.org<mailto:altoext-bounces@ietf.org>] on behalf of Tina TSOU [Tina.Tsou.Zouting@huawei.com<mailto:Tina.Tsou.Zouting@huawei.com>]
Sent: Sunday, March 25, 2012 5:14 AM
To: Vijay K. Gurbani; Spencer Dawkins
Cc: altoext@ietf.org<mailto:altoext@ietf.org>; alto@ietf.org<mailto:alto@ietf.org>
Subject: Re: [altoext] i2aex BOF - SDN Use Case

Attached are the updated slides after some discussion with chairs. Thank you.
C u soon in Paris.

Tina


> -----Original Message-----
> From: altoext-bounces@ietf.org<mailto:altoext-bounces@ietf.org> [mailto:altoext-bounces@ietf.org<mailto:altoext-bounces@ietf.org>] On
> Behalf Of Tina TSOU
> Sent: Friday, March 23, 2012 4:34 PM
> To: Vijay K. Gurbani; Spencer Dawkins
> Cc: altoext@ietf.org<mailto:altoext@ietf.org>; alto@ietf.org<mailto:alto@ietf.org>
> Subject: [altoext] i2aex BOF - SDN Use Case
>
> Dear i2aex BoF chairs,
> Attached please find slides for
>
> Infrastructure to Application Exposure (I2AEX) Use Case
>
>  - Software Define/Driven Networking (SDN)
>
>
> Comments are welcome.
>
>
> Tina
_______________________________________________
alto mailing list
alto@ietf.org<mailto:alto@ietf.org>
https://www.ietf.org/mailman/listinfo/alto