Re: [Lsr] Moving Forward [Re: Flooding Reduction Draft Redux]

[Peter Psenak <[e-mail]>]

Hi Huaimo,

On 13/02/2019 22:50 , Huaimo Chen wrote:
> Hi Peter,
>
>     My explanations/answers are in line below with prefix [HC].
>
> -----Original Message-----
> From: Peter Psenak [mailto:[e-mail]]
> Sent: Wednesday, February 6, 2019 4:58 AM
> To: Huaimo Chen <[e-mail]>; Acee Lindem (acee) <[e-mail]>; Christian Hopps <[e-mail]>; [e-mail]
> Subject: Re: [Lsr] Moving Forward [Re: Flooding Reduction Draft Redux]
>
> Hi Huaimo,
>
> On 03/02/2019 17:58 , Huaimo Chen wrote:
>> Hi Acee,
>>
>>
>>
>>     I agree with you on keeping the signaling for two modes. The other
>> parts for the distributed solution need to be removed.

optimized flooding is not only about algorithm to calculate the flooding 
topology and the way it is distributed/computed. It is also about local 
rules to make sure the flooding remains consistent. These are 
_independent_ of centralized/distributed modes. And it make no sense to 
specify these rules in two drafts.
>
> There are no "other" parts specific for the distributed solution.
>
> [HC] Some behaviors for the distributed solution/mode are described in draft-li-dynamic-flooding. For example, there are a few of places from page 27 to 30, which define the behaviors specific for the distributed solution/mode.

I strongly disagree. The fact that we say in centralized mode area 
leader recomputes and in distributed mode all nodes recompute make no 
difference in behavior.

thanks,
Peter

>
> draft-li-dyanmic-flooding defines:
>
> 1. the signalling that is common and used by both modes 2. distribution of the flooding-topology, which is specific to centralized mode 3. common behavior of the nodes that support the extension, which is independent of the mode of operation.
>
> [HC] In addition to these, draft-cc-lsr-flooding-reduction defines more, including concrete protections, operations, and algorithms for computing a flooding topology.
>
> Best Regards,
> Huaimo
>
> thanks,
> Peter
>
>
>>
>>
>>
>> Best Regards,
>>
>> Huaimo
>>
>> *From:* Acee Lindem (acee) [mailto:[e-mail]]
>> *Sent:* Sunday, February 3, 2019 11:45 AM
>> *To:* Huaimo Chen <[e-mail]>; Christian Hopps
>> <[e-mail]>; [e-mail]
>> *Subject:* Re: [Lsr] Moving Forward [Re: Flooding Reduction Draft
>> Redux]
>>
>>
>>
>> Hi Huaimo,
>>
>>
>>
>> See inline.
>>
>>
>>
>> *From: *Lsr <[e-mail] <mailto:[e-mail]>> on
>> behalf of Huaimo Chen <[e-mail]
>> <mailto:[e-mail]>>
>> *Date: *Saturday, February 2, 2019 at 12:27 AM
>> *To: *Christian Hopps <[e-mail] <mailto:[e-mail]>>,
>> "[e-mail] <mailto:[e-mail]>" <[e-mail]
>> <mailto:[e-mail]>>
>> *Subject: *Re: [Lsr] Moving Forward [Re: Flooding Reduction Draft
>> Redux]
>>
>>
>>
>> Hi Everyone,
>>
>>
>>
>> We proposed the distributed solution first, and Tony proposed the
>> centralized solution first. Tony added the distributed solution
>> (except for the algorithms to compute flooding topology) into his
>> draft. And then we added the centralized solution into our draft. The
>> latest versions of the two drafts have largely converged at least at
>> the high level to a solution for solving the same problem.
>>
>>
>>
>> Our draft has multiple key technical advantages over Tony's draft as
>> we described in our email to the LSR list, which are summarized below:
>>
>> 1.       It uses a fraction of flooding resource (i.e., it is multiple
>> times more efficient in flooding topology encoding);
>>
>> 2.       It provides fault tolerance to multiple failures, minimizing
>> impact on network convergence, thus minimizing traffic lose; and
>>
>> 3.       It is simpler and needs less processing time (i.e., faster and
>> more efficient) in multiple scenarios.
>>
>> Based on the technical merits, our draft should be moved forward.
>> However, Chair proposed to move Tony's draft forward and have us work
>> on a distributed algorithm as we started with.
>>
>>
>>
>> I think that the distributed solution in Tony's draft needs to be
>> removed and they work on the centralized solution. We remove the
>> centralized solution from our draft and work on the distributed solution.
>>
>>
>>
>> I'm against "cutting the baby in half" given that the signaling for
>> the distributed solution is a proper subset of what is required for
>> the centralized solution. It is undesirable to have different
>> signaling for the two modes. For the distributed algorithm you are
>> proposing, do see problems with the signaling?
>>
>>
>>
>> Thanks,
>>
>> Acee
>>
>>
>>
>> Best Regards,
>>
>> Huaimo
>>
>>
>>
>> -----Original Message-----
>>
>> From: Lsr [mailto:[e-mail]] On Behalf Of Christian Hopps
>>
>> Sent: Friday, February 1, 2019 7:26 AM
>>
>> To: [e-mail] <mailto:[e-mail]>
>>
>> Cc: [e-mail] <mailto:[e-mail]>
>>
>> Subject: [Lsr] Moving Forward [Re: Flooding Reduction Draft Redux]
>>
>>
>>
>>
>>
>> Summary of where we are at with dynamic flooding reduction:
>>
>>
>>
>> - We have a well written original work that came first and described
>> the problems as well as a TLVs to allow for a centralized solution
>> (draft-li-dyanmic-flooding). We do not need to standardize the
>> centralized algorithm.
>>
>>
>>
>> - A small change to this work allowed for distributed algorithms and
>> for outside work on distributed algorithms to continue in parallel.
>>
>>
>>
>> - We have another original work that started primarily as a
>> distributed algorithm
>>
>>    (draft-cc-ospf-flooding-reduction)
>>
>>
>>
>> - Finally we also have:
>>
>>    - Cross-pollination of ideas.
>>
>>    - Failed attempts at merging.
>>
>>    - An authors list "Arms-Race".
>>
>>
>>
>> Moving forward:
>>
>>
>>
>> - During IETF 103 I proposed we have no conflict if we:
>>
>>
>>
>>    1) adopt draft-li-lsr-dyanmic-flooding as the base WG document.
>>
>>    2) have authors of draft-cc-lsr-flooding-reduction work on a
>> distributed algorithm as they started with.
>>
>>
>>
>> - Acee agreed during the meeting (as chair) that this was the best way
>> forward. We had some agreement form the floor as well..
>>
>>
>>
>> - Any good ideas regarding the distribution of a centralized topology
>> can be debated and added (with appropriate attribution) to the base
>> document after we adopt one.
>>
>>
>>
>> - This is what happens when we adopt a document as WG work, we work on it.
>>
>>
>>
>> - The original authors of the distributed solution can continue to
>> work on their distributed algorithm in a separate document which would
>> also need standardization.
>>
>>
>>
>> Does anyone see a serious problem with this path forward?
>>
>>
>>
>> Thanks,
>>
>> Chris & Acee.
>>
>> LSR Chairs.
>>
>>
>>
>> Christian Hopps <[e-mail] <mailto:[e-mail]>> writes:
>>
>>
>>
>>> We've had the authors of the individual conflicting drafts take a
>>> shot
>> at merging their work.
>>
>>>
>>
>>>    This has failed.
>>
>>>
>>
>>> Here is the full history (which I also summarized during IETF103 as
>> well). I will send a second email discussing this.
>>
>>>
>>
>>> - Jan 2, 2018 Publication: draft-li-dynamic-flooding and
>> drfat-li-dynamic-flooding-isis
>>
>>>   published centralized solution.
>>
>>>
>>
>>> - Mar 5, 2018 Publication: draft-cc-isis-flooding-reduction and
>> draft-cc-ospf-flooding-reduction
>>
>>>   published distributed solution.
>>
>>>   - mention of centralized solution asserting it is not good choice.
>>
>>>
>>
>>> - IETF 101 (Mar 2018)
>>
>>>   - Video:
>> https://www.youtube.com/watch?v=qHmT4ytMn4w&list=PLC86T-6ZTP5j_HaBNdfP
>> bgxGIp22cnaWS
>>
>>>   - Minutes:
>> https://datatracker.ietf.org/meeting/101/materials/minutes-101-lsr-00
>>
>>>   - draft-li-dynamic-flooding-02 presented (1 author). at IETF 101
>>
>>>     - Generally well received.
>>
>>>   - draft-cc-ospf-flooding-reduction-00 (4 authors) presented.
>>
>>>     - Serious problems immediately found during presentation -- not
>> fully baked.
>>
>>>
>>
>>> - Mar 18, 2018 draft-li-dynamic-flooding-03 published (1 author)
>>
>>> - Mar 27, 2018 draft-li-dynamic-flooding-04 published (1 author)
>>
>>> - Apr 20, 2018 draft-cc-ospf-flooding-reduction-01 revised
>>
>>> - Jun 28, 2018 draft-li-dynamic-flooding-05 published (2 authors)
>>
>>>   - *SMALL CHANGE TO SUPPORT DISTRIBUTED ALGORITHM*.
>>
>>>   - Does not specify distributed algorithm only how to indicate one
>>> in
>> use, small change.
>>
>>>
>>
>>> - Jul 2, 2018 draft-cc-ospf-flooding-reduction-02 published
>>
>>>
>>
>>> - IETF 102 (Jul 14, 2018)
>>
>>>   - draft-li-dynamic-flooding-05 presented.
>>
>>>   - draft-cc-ospf-flooding-reduction-02 presented.
>>
>>>
>>
>>> - Sep 12, 2018 draft-cc-ospf-flooding-reduction-03 (4 authors)
>>
>>>   - *LARGE CHANGE ADDS NEW CENTRALIZED SOLUTION*.
>>
>>>
>>
>>> - Sep 20, 2018 draft-cc-ospf-flooding-reduction-04 (4 authors)
>>
>>>
>>
>>> - Oct 21, 2018 draft-li-lsr-dynamic-flooding-00 and -01 (5 authors)
>>
>>>
>>
>>> - IETF 103 (Nov 3, 2018)
>>
>>>
>>
>>>   - Chairs give direction
>>
>>>
>>
>>>     - draft-li-lsr-dynamic-flooding-05 having come first, being well
>> written and not
>>
>>>       specifying a distributed algorithm (merely allowing for one) is
>> the correct vehicle
>>
>>>       to adopt as a base document.
>>
>>>
>>
>>>     - Distributed algorithm work (the original basis for
>> draft-cc-ospf-flooding-reduction)
>>
>>>       should continue as a separate document form the base which
>>> would
>> thus we have no
>>
>>>       conflicts.
>>
>>>
>>
>>> - In the meantime the authors try and merge work, this fails.
>>
>>>
>>
>>> - Dec 3, 2018 draft-li-lsr-dynamic-flooding-02 (7 authors)
>>
>>>
>>
>>> - Dec 10, 2018 draft-cc-lsr-flooding-reduction-00 (4 authors)
>>
>>>
>>
>>> - Jan 7, 2019  draft-cc-lsr-flooding-reduction-01 (8 authors)
>>
>>
>>
>>
>>
>> _______________________________________________
>> Lsr mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/lsr
>>
>
> .
>