draft-iab-raws-report-00.txt posted to the ID repository

[David Meyer]

       
        Folks,

        First, sorry about the wide distribution and any
        duplicates you may receive.

        I just posted draft-iab-raws-report-00.txt posted to the
        ID repository. Until it gets there, I've put a copy on

         http://www.1-4-5.net/~dmm/draft-iab-raws-report-00.txt

        for your light holiday reading :-)

        Looking forward to your comments and the ongoing
        discussion.

        --dmm (for dmm, lixia, and kevin)



_______________________________________________
Architecture-discuss mailing list
Architecture-discuss@...
https://www1.ietf.org/mailman/listinfo/architecture-discuss

[David Ward-2]

Dave -

    As editor of the document that captures the presentations and discussion
during this workshop, how should incorrect statements be handled?
Unfortunately, there are some errors in the statements that don't quite
reflect reality (once example is in the Routing convergence section 7.3).
Since the doc represents what was discussed, it may not be appropriate to
correct the statements. Would any corrections or clarifications go into the
future problem statement document that you mention?

Thanks

-DWard


On 12/15/06 9:37 AM, "David Meyer" <dmm@...> wrote:

>
> Folks,
>
> First, sorry about the wide distribution and any
> duplicates you may receive.
>
> I just posted draft-iab-raws-report-00.txt posted to the
> ID repository. Until it gets there, I've put a copy on
>
> http://www.1-4-5.net/~dmm/draft-iab-raws-report-00.txt
>
> for your light holiday reading :-)
>
> Looking forward to your comments and the ongoing
> discussion.
>
> --dmm (for dmm, lixia, and kevin)
>
> _______________________________________________
> RAM mailing list
> RAM@...
> https://www1.ietf.org/mailman/listinfo/ram

_______________________________________________
Architecture-discuss mailing list
Architecture-discuss@...
https://www1.ietf.org/mailman/listinfo/architecture-discuss

[David Meyer]

        Dave,

>     As editor of the document that captures the presentations and discussion
> during this workshop, how should incorrect statements be handled?
> Unfortunately, there are some errors in the statements that don't quite
> reflect reality (once example is in the Routing convergence section 7.3).
> Since the doc represents what was discussed, it may not be appropriate to
> correct the statements. Would any corrections or clarifications go into the
> future problem statement document that you mention?

        I understand your problem with 7.3. Clearly, we
        understand that IGPs can (and do) converge more quickly
        that BGP does. If I understand your concern, it is with
        this text (please correct me if I'm wrong):

          The IGP convergence can also be very slow, which can
          lead to intolerable performance problems for real time
          applications such as VoIP.  The cause for this slow
          convergence can be due to multiple factors, including...

        in that it seems to suggest that IGP convergence is
        *slower* that BGP convergence (correct)? If so, will
        text like:

           While IGPs are designed to and do converge much more
           qucikly that BGP might, the workshop participants were
           concerened that the growth of the size of the DFZ RIB,
           in addition to the various specical purpose routes
           they must carry can slow  IGP convergence. The cause
           for this slow convergence can be due to multiple
           factors, including....

        or similar address your concern?

        thnx for the comments,

        --dmm



_______________________________________________
Architecture-discuss mailing list
Architecture-discuss@...
https://www1.ietf.org/mailman/listinfo/architecture-discuss

[David Ward-2]

Dave -

    It is better but, still not completely correct (given the remainder of
the section). The meta-point is that the doc is a statement of what was
discussed in the workshop. There are factual errors in the doc in addition
to what I cited (I purposefully chose perhaps the least controversial in
7.3). Your answer below implies that you would like the workshop doc to be
correct vs leaving it as a statement from the workshop and then having the
problem statement fully clarified. I suggest that you proceed forward w/ the
problem statement and not potentially revise history. Of course, it is up to
authors and group ...

0.02$

-DWard


On 12/15/06 10:21 AM, "David Meyer" <dmm@...> wrote:

> Dave,
>
>>     As editor of the document that captures the presentations and discussion
>> during this workshop, how should incorrect statements be handled?
>> Unfortunately, there are some errors in the statements that don't quite
>> reflect reality (once example is in the Routing convergence section 7.3).
>> Since the doc represents what was discussed, it may not be appropriate to
>> correct the statements. Would any corrections or clarifications go into the
>> future problem statement document that you mention?
>
> I understand your problem with 7.3. Clearly, we
> understand that IGPs can (and do) converge more quickly
> that BGP does. If I understand your concern, it is with
> this text (please correct me if I'm wrong):
>
>  The IGP convergence can also be very slow, which can
>  lead to intolerable performance problems for real time
>  applications such as VoIP.  The cause for this slow
>  convergence can be due to multiple factors, including...
>
> in that it seems to suggest that IGP convergence is
> *slower* that BGP convergence (correct)? If so, will
> text like:
>
>   While IGPs are designed to and do converge much more
>   qucikly that BGP might, the workshop participants were
>   concerened that the growth of the size of the DFZ RIB,
>   in addition to the various specical purpose routes
>   they must carry can slow  IGP convergence. The cause
>   for this slow convergence can be due to multiple
>   factors, including....
>
> or similar address your concern?
>
> thnx for the comments,
>
> --dmm
>

_______________________________________________
Architecture-discuss mailing list
Architecture-discuss@...
https://www1.ietf.org/mailman/listinfo/architecture-discuss

[David Meyer]

On Fri, Dec 15, 2006 at 10:44:21AM -0600, David Ward wrote:

> Dave -
>
>     It is better but, still not completely correct (given the remainder of
> the section). The meta-point is that the doc is a statement of what was
> discussed in the workshop. There are factual errors in the doc in addition
> to what I cited (I purposefully chose perhaps the least controversial in
> 7.3). Your answer below implies that you would like the workshop doc to be
> correct vs leaving it as a statement from the workshop and then having the
> problem statement fully clarified. I suggest that you proceed forward w/ the
> problem statement and not potentially revise history. Of course, it is up to
> authors and group ...

        Well, what the report is supposed to reflect is what
        happened at the workshop. So I didn't intend to revise
        history, but rather clarify what was said.

        --dmm


_______________________________________________
Architecture-discuss mailing list
Architecture-discuss@...
https://www1.ietf.org/mailman/listinfo/architecture-discuss

[HeinerHummel]

I am glad that this IAB-workshop confirms my saying that there is no future with IPv6 and BGP.Both are children of the 90's. Also: it is not only the bad convergency of BGP (which is due to its flat design and which  is also the cause  for instability(=routing churn) AND non-scalability. Furthermore it prevents real-time traffic balancing for good. OSPF could do better, but doesn't either as long as there is this SPF-dogma ( I accused it four years ago in a longer thread "Rambo-SPF....") where each router behaves as if it were the center of the internet :-(

 

Heiner

 

 

In einer eMail vom 15.12.2006 18:04:48 Westeuropäische Normalzeit schreibt dmm@...:

On Fri, Dec 15, 2006 at 10:44:21AM -0600, David Ward wrote:

> Dave -
>
>     It is better but, still not completely correct (given the remainder of
> the section). The meta-point is that the doc is a statement of what was
> discussed in the workshop. There are factual errors in the doc in addition
> to what I cited (I purposefully chose perhaps the least controversial in
> 7.3). Your answer below implies that you would like the workshop doc to be
> correct vs leaving it as a statement from the workshop and then having the
> problem statement fully clarified. I suggest that you proceed forward w/ the
> problem statement and not potentially revise history. Of course, it is up to
> authors and group ...

    Well, what the report is supposed to reflect is what
    happened at the workshop. So I didn't intend to revise
    history, but rather clarify what was said.

    --dmm

_______________________________________________
Architecture-discuss mailing list
Architecture-discuss@...
https://www1.ietf.org/mailman/listinfo/architecture-discuss

 

_______________________________________________
Architecture-discuss mailing list
Architecture-discuss@...
https://www1.ietf.org/mailman/listinfo/architecture-discuss