Re: [Supa] draft-bi-supa-gap-analysis

["Susan Hares" <[e-mail]>]

Tina:

 

Just to make sure I am clear.  The I2RS is chartered to provide dynamic (I2rs)  protocol independent models (E.g. RIB, topology, and PBR).  This includes a generic topology model for protocol independent models.

 

Sue 

 

From: Tina TSOU [mailto:[e-mail]] 
Sent: Monday, March 02, 2015 10:08 AM
To: Susan Hares; 'Romascanu, Dan (Dan)'; 'Ying Cheng'; 'Dacheng Zhang'
Cc: 'Marc Blanchet'; 'Jean-Fran?ois Tremblay'; 'Natale, Bob'; [e-mail]; 'Jun Bi'
Subject: RE: [Supa] draft-bi-supa-gap-analysis

 

Dear all,

 

Great discussion.

 

I edited a little bit on Ying’s figure as attached, for discussion purpose.

 

As discussed in this thread, at this point in time, the existing WG charters do not specify which WG is in charge to provide a generic toplogy model and that we are happy to work with any WG that might get chartered to provide the generic topology model.

 

 

Thank you,

Tina

 

From: Susan Hares [mailto:[e-mail]] 
Sent: Monday, March 02, 2015 10:51 PM
To: 'Romascanu, Dan (Dan)'; 'Ying Cheng'; 'Dacheng Zhang'; Tina TSOU
Cc: 'Marc Blanchet'; 'Jean-Fran?ois Tremblay'; 'Natale, Bob'; [e-mail]; 'Jun Bi'
Subject: RE: [Supa] draft-bi-supa-gap-analysis

 

Dan, Ying, and Dacheng: 

 

This model seems to really mix many pieces of the work.  SUPA’s problem statement claims to look at the service level for the configuration.  If so, the work is either configuration (netmod/netconf)?  If it is dynamic, the protocol independent topologies and policy for services are being handled by the I2RS work.  

 

•   draft-chen-i2rs-topology-im-00.txt <http://datatracker.ietf.org/doc/draft-clemm-i2rs-yang-network-topo/>   (summary of PIT topologies including service topologies, pending)   

•   draft-clemm-i2rs-yang-network-topo-02 <http://datatracker.ietf.org/doc/draft-clemm-i2rs-yang-network-topo/>  (generic model) 

•   Draft-dong-i2rs-inventory-00.txt    (generic model addition for resources, pending) 

•    draft-zhang-i2rs-l1-topo-yang-model-00 <http://datatracker.ietf.org/doc/draft-zhang-i2rs-l1-topo-yang-model/> 

•   draft-dong-i2rs-l2-network-topology-00 <http://datatracker.ietf.org/doc/draft-dong-i2rs-l2-network-topology/> 

•   draft-clemm-i2rs-yang-l3-topo-00 <http://datatracker.ietf.org/doc/draft-clemm-i2rs-yang-l3-topo/>   

•   draft-dong-i2rs-l2-network-topology-00 <http://datatracker.ietf.org/doc/draft-dong-i2rs-l2-network-topology/> 

•   Draft-wang-i2rs-service-topo-dm-00 <http://tools.ietf.org/html/draft-wang-i2rs-yang-service-topo-dm-00>    - exist, revision pending 

•   Draft-wang-i2rs-service-topo-sfc-dm-00 (service model) – exists , revision pending 

•   Draft-dunbar-i2rs-discover-traffic-rules-00 (policy models) - pending

 

If SUPA is dealing with intent, it is not covered in the problem statement or the use cases.  If you are adding intent at this late date, you should consider the intent-specific work in Nemo ([e-mail] work (www.nemo-project.org) and the service topology). 

 

Can you tell me why you are taking this approach to SUPA? This diagram changes the SUPA work to middle box area without providing any support for the topology? 

 

Sue 

 

From: Supa [mailto:[e-mail]] On Behalf Of Romascanu, Dan (Dan)
Sent: Monday, March 02, 2015 7:43 AM
To: Romascanu, Dan (Dan); Ying Cheng; Dacheng Zhang; Tina TSOU
Cc: Marc Blanchet; Jean-Fran?ois Tremblay; Natale, Bob; [e-mail]; Jun Bi
Subject: Re: [Supa] draft-bi-supa-gap-analysis

 

One more thing that I am missing here is the topology input. As things look now we will import different layers of topology from places like I2RS or TEAS – we may need to represent these. 

 

Regards,

 

Dan

 

 

From: Supa [mailto:[e-mail]] On Behalf Of Romascanu, Dan (Dan)
Sent: Monday, March 02, 2015 1:53 PM
To: Ying Cheng; Dacheng Zhang; Tina TSOU
Cc: Marc Blanchet; Jun Bi; Jean-Fran?ois Tremblay; [e-mail]; Natale, Bob
Subject: Re: [Supa] draft-bi-supa-gap-analysis

 

Questions (if there are references already some place else please show me where they are: 

-          What ‘Intent’ means in the figure? 

-          I assume the middle box network element (NE) is labelled Deploy and not Depoly. What is the difference between it and the other two similar boxes with labels ‘Active’? 

-          What is the functionality of the Service Management box? Only service creation or more? 

Thanks and Regards,

Dan

 

 

From: Supa [mailto:[e-mail]] On Behalf Of Ying Cheng
Sent: Saturday, February 28, 2015 9:06 AM
To: Dacheng Zhang; Tina TSOU; Romascanu, Dan (Dan)
Cc: Jun Bi; Jean-François Tremblay; Natale, Bob; [e-mail]; Marc Blanchet
Subject: Re: [Supa] draft-bi-supa-gap-analysis

 

Hi all,

 

Based upon Dacheng's texts on SUPA scope, it is proposed to add the following figure as the starting point for our further discussion.  

 

Any comments are welcomed and appreciated.

 

BR,

Ying

 



 

 

2015-02-28 

  _____  

Ying CHENG  程莹
China Unicom Network Technology Research Institute 中国联通网络技术研究院
Tel: [phone]-7702
Mobile: [phone]

  _____  

发件人： Dacheng Zhang 

发送时间： 2015-02-27  22:26:06 

收件人： Tina TSOU; Romascanu, Dan (Dan) 

抄送： Ying Cheng(联通集团联通网络技术研究院); Jun Bi; Jean-François Tremblay; Marc Blanchet; Natale, Bob; [e-mail] 

主题： Re: [Supa] draft-bi-supa-gap-analysis 

SUPA Scope and Focus：

SUPA will focus in the first phase on inter-datacenter traffic management in the use case of a distributed data center, including the automated provisioning of site-to-site virtual private networks of various types, including service configuratioin model ,generic topology model and generic network-level policy model. All of them describe network behaviors for a service. Specially, the existing WG charters do not specify which WG is in charge to provide a generic toplogy model and that SUPA is happy to work with any WG that might get chartered to provide the generic topology model.

 

Relationship with I2RS:

I2RS defines a collection of protocol-based control or management

interfaces for route system；

SUPA can leverage the exsiting protocol and models defined I2RS;

 

Relationship with BESS:

BESS defines, specifies and extends network services based on BGP and sets up VPNs by configurating BGP;

SUPA doesn't care how to set up VPNs by some specific protocol;

SUPA can leverage the exsiting protocol and models defined BESS;

 

Relationship with OpenStack Congress:             

Congress is intended to be a general-purpose, high-level policy component of OpenStack.  To enforce the policies across compute, storage and networking, Congress would carve off subpolicies and send them to the appropriate enforcement point in the cloud, Supa could be one of them in the networking domain.

 

Relationship with TEAS:

TEAS is responsible for defining MPLS and GMPLS traffic engineering architecture, standardizing the RSVP-TE signaling protocol, and identifying required related control-protocol functions.

SUPA is responsible for defining ECA policy model to implement TE.

 

Relationship with I2nsf:

The relation or intersection between the SUPA and I2NSF may be in the information and data model for policies. I2NSF does not have or care about topologies.

 

发件人: Tina TSOU <[e-mail]>
日期: 2015年2月26日 星期四 下午11:18
至: "Romascanu, Dan (Dan)" <[e-mail]>
抄送: Ying Cheng <[e-mail]>, Jun Bi <[e-mail]>, Jean-François Tremblay <[e-mail]>, Marc Blanchet <[e-mail]>, "Natale, Bob" <[e-mail]>, "[e-mail]" <[e-mail]>
主题: Re: [Supa] draft-bi-supa-gap-analysis

 

Dear Dan et al,

 

Good point.

 

I would suggest the following template to facilitate gap analysis discussion? Proposals? Let's see who fills it quickly;)

Ideally, there should be a figure.

SUPA scope and focus: xxx

Relationship with ir2s: xxx

Relationship with teas: xxx

Relationship with OpenStack Congress: xxx

Relationship with ODL NIC: xxx

 

 

Thank you,

Tina


On Feb 25, 2015, at 6:08 PM, Romascanu, Dan (Dan) <[e-mail]> wrote:

<!--[if mso 9]--> <!--[endif]--> 

I agree that for the purpose of the discussions at the BOF slides are useful. However, the IETF mode of work gives preference to Internet-Drafts, and when chartering will be considered ‘are there relevant Internet-Drafts?’ is one of the questions the IESG will ask. My experience is that the IESG prefers I-Ds to powerpoint. The current I-D on gaps analysis was not updated since last September and for somebody searching and reading I-Ds the reading of this document can be mis-leading about where the team is in the discussions. 

 

Regards,

 

Dan

 

 

From: Supa [mailto:[e-mail]] On Behalf Of Natale, Bob
Sent: Wednesday, February 25, 2015 8:45 AM
To: Ying Cheng; Marc Blanchet; Tina Tsou
Cc: Jun Bi; Jean-François Tremblay; [e-mail]
Subject: Re: [Supa] draft-bi-supa-gap-analysis

 

+1 … should be relatively easy to represent visually and, once agreed upon, would form the basis of the corresponding I-D.

 

Avanti,

BobN

 

From: Supa [mailto:[e-mail]] On Behalf Of Ying Cheng
Sent: Wednesday, February 25, 2015 12:59 AM
To: Marc Blanchet; Tina Tsou
Cc: Jun Bi; Jean-François Tremblay; [e-mail]
Subject: Re: [Supa] draft-bi-supa-gap-analysis

 

Hi all,

 

I agree with Marc's suggestion. Slides of gap analysis with figures are more helpful for discussion.

 

BR,

Ying

 

 

2015-02-25 


  _____  


Ying CHENG  程莹
China Unicom Network Technology Research Institute 中国联通网络技术研究院
Tel: [phone]-7702
Mobile: [phone]


  _____  


发件人： Marc Blanchet 

发送时间： 2015-02-25  00:21:24 

收件人： Tina Tsou 

抄送： Jun Bi; Jean-François Tremblay; [e-mail] 

主题： Re: [Supa] draft-bi-supa-gap-analysis 

 

Le 2015-02-23 à 23:43, Tina TSOU <[e-mail]> a écrit :

 

Dear Jun and JF, 

 

Per this Monday's meeting, the gap analysis draft is one of the drafts that must be updated by Mar 9.

 

http://datatracker.ietf.org/doc/draft-bi-supa-gap-analysis/ <https://urldefense.proofpoint.com/v2/url?u=http-3A__datatracker.ietf.org_doc_draft-2Dbi-2Dsupa-2Dgap-2Danalysis_&d=AwMGaQ&c=BFpWQw8bsuKpl1SgiZH64Q&r=I4dzGxR31OcNXCJfQzvlsiLQfucBXRucPvdrphpBsFA&m=75NnaKfHSnJEyTvLp9a5KStGqK_KqY0aUr8JpCuvi2A&s=d_YdL_lV5dqUgzvEw-Y8PvZYpUd4Q0aB-mAZj_0niGc&e=> 

 

Would you work on it?

 

I remember you signed up to work on this draft. I may miss the other people who also signed up for this. Pls team up with Jun and JF.

 

Dear Tina,

 

I see two kind of gap analysis:

a) a full blown internet draft with all the details of everything

b) one slide or two that describes where each wg is shown and what is supa would be doing. I could even see some graphic representation.

 

The overall goal is to  have a good discussion during the BOF.  From that perspective, I think we should work on b), not a).

 

my 2 cents.

 

Marc.

 

 

 

Thank you,

Tina

_______________________________________________
Supa mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/supa <https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ietf.org_mailman_listinfo_supa&d=AwMGaQ&c=BFpWQw8bsuKpl1SgiZH64Q&r=I4dzGxR31OcNXCJfQzvlsiLQfucBXRucPvdrphpBsFA&m=75NnaKfHSnJEyTvLp9a5KStGqK_KqY0aUr8JpCuvi2A&s=xtw8PCJw-uH3Na6nOv2ntoPkGf7N9UkqKDggiMM681s&e=> 

 

如果您错误接收了该邮件，请通过电子邮件立即通知我们。 <mailto:请回复邮件到[e-mail]，即可以退订此邮件。我们将立即将您的信息从我们的发送目录中删除> 请回复邮件到[e-mail]，即可以退订此邮件。我们将立即将您的信息从我们的发送目录中删除。 If you have received this email in error please notify us immediately by e-mail. Please reply to [e-mail] if you want to unsubscribe from this mail. We will immediately remove your information from our sending catalogue.

_______________________________________________ Supa mailing list [e-mail] https://www.ietf.org/mailman/listinfo/supa <https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ietf.org_mailman_listinfo_supa&d=AwMGaQ&c=BFpWQw8bsuKpl1SgiZH64Q&r=I4dzGxR31OcNXCJfQzvlsiLQfucBXRucPvdrphpBsFA&m=8tlXfMBYiyXdl3QX_uaZEwaErzoCYOw1TwelXmRq-5M&s=hklOUyvfQM5mQ7KybkMJTKMVTlnu7lUcIuXQYHWTCgE&e=>  

如果您错误接收了该邮件，请通过电子邮件立即通知我们。请回复邮件到[e-mail]，即可以退订此邮件。我们将立即将您的信息从我们的发送目录中删除。 If you have received this email in error please notify us immediately by e-mail. Please reply to [e-mail] if you want to unsubscribe from this mail. We will immediately remove your information from our sending catalogue.