Hi Donald,

Thanks for the review. A few comments below...

On 17/04/2013 07:08, Donald Eastlake wrote:
> Hi,
> 
> I think the draft is a good documentation of the issues and the
> benefits of using the IPv6 flow ID with load balancing.
> 
> It would be more consistent with the completeness of the rest of the
> document to have a more complete list of typical load balancer server
> selection algorithms in the last paragraph of Section 5.

That is not really "on the wire" stuff though - isn't it sufficient to
indicate that there are multiple methods?

> There are people who believe that Informational documents can't have
> normative references. I am generally inclined that way myself.

There doesn't seem to be a rule about that. Coincidentally, I'm
in the middle of an argument about that over on apps-discuss, where
we've been asked to promote some Informational documents to normative
references in an Informational RFC. I think that's silly. I'd be happy
to drop the distinction here.

> Various descriptions in the draft, like "statistically rare" could be
> made a bit more quantitative. I'm not saying a detailed numeric
> analysis is needed but a rough number based on reasonable simplifying
> assumptions might help to make things a little less fuzzy.

Yes, that is worth some thought. At least for flow label collisions,
it's discussed in the flow label spec (hence our conversation some
time ago about hash algorithms and FNV). We may be able to cover this
by reference.

Regards,

    Brian

> 
> Those are all my comments.
> 
> Thanks,
> Donald
> =============================
>  Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
>  155 Beaver Street, Milford, MA 01757 USA
>  [email protected]
> 
> 
> On Tue, Apr 16, 2013 at 2:57 AM, Sheng Jiang <[email protected]> wrote:
>> Hi, dear Donald,
>>
>> You agreed to review our flow label balancing document during the IntArea 
>> meeting in Orlando. The blow is the link for our document. Could you please 
>> review it and give your comments?
>>
>> http://tools.ietf.org/html/draft-ietf-intarea-flow-label-balancing-00
>>
>> Many thanks and best regards,
>>
>> Sheng
> 
_______________________________________________
Int-area mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/int-area

Reply via email to