Re: [spring] New Version Notification for draft-dukes-spring-srv6-overhead-analysis-00.txt

[Ron Bonica <[e-mail]>]

Hello friend,

SRv6 and SRm6 contain the same information in their FIBs. The difference is that the FIBs are structured differently.

Are you assuming that a change in FIB structure constitutes a new forwarding plane. Can you find any text in the IPv6 specifications that support this assumption?

Exploring the other side of the coin, the RFC 4291 definition of an IPv6 address is very different from draft-filsfilscheng-spring-srv6-srh-comp-sl-enc definition of a C-SID container. Yet both can be placed in the destination address field of an IPv6 header. Is this still IPv6, or have we created a new data plane?

                                                                 Ron




Juniper Business Use Only
From: Vasilenko Eduard <[e-mail]>
Sent: Tuesday, June 30, 2020 3:21 AM
To: Ron Bonica <[e-mail]>; Darren Dukes (ddukes) <[e-mail]>; SPRING WG <[e-mail]>
Subject: RE: New Version Notification for draft-dukes-spring-srv6-overhead-analysis-00.txt

[External Email. Be cautious of content]

Hi Guru,
In my humble opinion statement that something compressed is not a new data plane could be a little wrong.
Genesis of the solution (all solutions comes from extension capabilities of IPv6) could only permit to tell that it is "the same protocol", or more precise "extension of the same protocol".
But if new forwarding table (with shorter labels) should be created in PFE - then it is probably new data plane. It occupies separate resources in data plane.
Eduard
From: spring [mailto:[e-mail]] On Behalf Of Ron Bonica
Sent: 30 июня 2020 г. 5:59
To: Darren Dukes (ddukes) <ddukes=[e-mail]<mailto:ddukes=[e-mail]>>; SPRING WG <[e-mail]<mailto:[e-mail]>>
Subject: Re: [spring] New Version Notification for draft-dukes-spring-srv6-overhead-analysis-00.txt

Darren,

The 6man WG is responsible for exactly one data plane. That is, IPv6.

IPv6 is extensible. So, new Routing headers and Options can be defined  without creating a new data plane. That's the whole idea of IPv6 extensibility. Do you believe new data planes were created when the following were defined:


  *   The NIMROD Routing header
  *   They Type 2 Routing header
  *   The RPL Routing header
  *   The Segment Routing Header

Also, the bullets under discussion have little to do with an "SRv6 Network Programming Overhead Analysis".

If your real goal is holistic comparison of SRm6 and draft-filfilscheng-srv6-srh-comp-sl-eng, I would be glad to work with you on that. But a holistic comparison should probably have representation from both sides. Otherwise, it is mere marketing.

                                                    Ron



Juniper Business Use Only
From: Darren Dukes (ddukes) <ddukes=[e-mail]<mailto:ddukes=[e-mail]>>
Sent: Monday, June 29, 2020 9:33 PM
To: Ron Bonica <[e-mail]<mailto:[e-mail]>>; SPRING WG <[e-mail]<mailto:[e-mail]>>
Subject: Re: New Version Notification for draft-dukes-spring-srv6-overhead-analysis-00.txt

[External Email. Be cautious of content]

Hi Ron. Thanks for reading the document.

You say about section 5:
> Also, many of the things that you say in that
> bullet list are blatantly false. For example,
> SRm6 does not introduce a new data plane.
> In is extremely orthodox IPv6.

SRm6 uses IPv6 for transport and it introduces a dataplane that maps 32 or 16 bit identifiers to a behavior and IPv6 address. Every SRm6 node must use this new dataplane implementation.  This is correct.

I believe the bullets I listed are relevant when holistically considering the proposals.

Sincerely,
  Darren


________________________________
From: Ron Bonica <rbonica=[e-mail]<mailto:rbonica=[e-mail]>>
Sent: Saturday, June 27, 2020 7:44:47 PM
To: Darren Dukes (ddukes) <[e-mail]<mailto:[e-mail]>>; SPRING WG <[e-mail]<mailto:[e-mail]>>
Subject: RE: New Version Notification for draft-dukes-spring-srv6-overhead-analysis-00.txt


Darren,



Your draft purports to be an "SRv6 Network Programming Overhead Analysis".  As such, it should address overhead analysis and avoid:



  *   Topics that are orthogonal to overhead analysis
  *   The appearance of attempting to position one compression strategy over another for reasons other than overhead



So, I recommend that you make the following changes to Section 5:



  *   The sentence "The mapping proposal, [I-D.bonica-spring-sr-mapped-six], does not bring any compression benefit compared to SRv6-native compression methods [I-D.filsfilscheng-spring-srv6-srh-comp-sl-enc]" gives the appearance of author bias. Please replaces it with a neutral sentence like, "The two  proposals [I-D.filsfilscheng-spring-srv6-srh-comp-sl-enc] [I-D.bonica-spring-sr-mapped-six] provide similar compression.
  *   You say that " The SRm6 proposal does have several deficiencies however, including:".... None of these have anything to do with overhead analysis. They don't belong in this document.
  *   Also, many of the things that you say in that bullet list are blatantly false. For example, SRm6 does not introduce a new data plane. In is extremely orthodox IPv6.



                                                                                                                        Ron









Juniper Business Use Only

From: spring <[e-mail]<mailto:[e-mail]>> On Behalf Of Darren Dukes (ddukes)
Sent: Saturday, June 27, 2020 1:22 AM
To: SPRING WG <[e-mail]<mailto:[e-mail]>>
Subject: [spring] Fw: New Version Notification for draft-dukes-spring-srv6-overhead-analysis-00.txt



[External Email. Be cautious of content]



Hello SPRING working group



There has been lots of work done in SPRING to develop, combine and refine methods of reducing the overhead of the SRv6 SRH.  We have some good submissions of requirements, framework analysis but no direct comparisons of different methods.



This draft kicks off that conversation with a simple analysis comparing SRv6 native compression available via draft-filsfilscheng-spring-srv6-srh-comp-sl-enc vs the mapped SRm6 proposal draft-bonica-spring-sr-mapped-six.





Thanks

  Darren





________________________________

From: [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Sent: Saturday, June 27, 2020 1:00 AM
To: Darren Dukes (ddukes) <[e-mail]<mailto:[e-mail]>>
Subject: New Version Notification for draft-dukes-spring-srv6-overhead-analysis-00.txt



A new version of I-D, draft-dukes-spring-srv6-overhead-analysis-00.txt
has been successfully submitted by Darren Dukes and posted to the
IETF repository.

Name:           draft-dukes-spring-srv6-overhead-analysis
Revision:       00
Title:          SRv6 Network Programming Overhead Analysis
Document date:  2020-06-27
Group:          Individual Submission
Pages:          9
URL:            https://www.ietf.org/internet-drafts/draft-dukes-spring-srv6-overhead-analysis-00.txt<https://urldefense.com/v3/__https:/www.ietf.org/internet-drafts/draft-dukes-spring-srv6-overhead-analysis-00.txt__;!!NEt6yMaO-gk!XQ30Nxg22KtJi8ikwlkNqwrZPkBjF4IixztGJYkMVwpqxc8PEt_xKQSRjEj0Z_T4$>
Status:         https://datatracker.ietf.org/doc/draft-dukes-spring-srv6-overhead-analysis/<https://urldefense.com/v3/__https:/datatracker.ietf.org/doc/draft-dukes-spring-srv6-overhead-analysis/__;!!NEt6yMaO-gk!XQ30Nxg22KtJi8ikwlkNqwrZPkBjF4IixztGJYkMVwpqxc8PEt_xKQSRjOMg93gm$>
Htmlized:       https://tools.ietf.org/html/draft-dukes-spring-srv6-overhead-analysis-00<https://urldefense.com/v3/__https:/urldefense...com/v3/__https:/urldefense...com/v3/__https:/tools.ietf..org/html/draft-dukes-spring-srv6-overhead-analysis-00__;!!NEt6yMaO-gk!XQ30Nxg22KtJi8ikwlkNqwrZPkBjF4IixztGJYkMVwpqxc8PEt_xKQSRjH9AbgoR$__;!!NEt6yMaO-gk!QrTNpOov227mx78vk0QW9dUiezD-3jNSo7hrwU5wJzCSX8FX6z6L4BIiAYtsAIIb$__;!!NEt6yMaO-gk!RxsuHx0YNYzQXzJ2rt5ZSZGFNHR3iYYUrPbwro_OkcXlb76SiuIv7eLzJm4pyMMH$>
Htmlized:       https://datatracker.ietf.org/doc/html/draft-dukes-spring-srv6-overhead-analysis<https://urldefense.com/v3/__https:/urldefense...com/v3/__https:/urldefense...com/v3/__https:/datatracker.ietf.org/doc/html/draft-dukes-spring-srv6-overhead-analysis__;!!NEt6yMaO-gk!XQ30Nxg22KtJi8ikwlkNqwrZPkBjF4IixztGJYkMVwpqxc8PEt_xKQSRjOjtZl8r$__;!!NEt6yMaO-gk!QrTNpOov227mx78vk0QW9dUiezD-3jNSo7hrwU5wJzCSX8FX6z6L4BIiAXN2pMyI$__;!!NEt6yMaO-gk!RxsuHx0YNYzQXzJ2rt5ZSZGFNHR3iYYUrPbwro_OkcXlb76SiuIv7eLzJvAzSORN$>


Abstract:
   SRv6 network programming provides the framework for the best
   compression of an IPv6 header within an SR domain.  This document
   provides the analysis to illustrate this fact.


The IETF Secretariat