Re: [tsvwg] [iccrg] New Internet Draft: Congestion Signaling (CSIG)

On 23/02/2024 07:18, Sebastian Moeller wrote:
> Hi Tom,
>
>
>> On 23. Feb 2024, at 02:44, Tom Herbert <[email protected]> wrote:
>>
>> On Thu, Feb 22, 2024 at 12:36 AM Sebastian Moeller <[email protected]> wrote:
>>> Hi Jai,
>>>
>>>
>>>> On 21. Feb 2024, at 17:06, Jai Kumar <[email protected]> wrote:
>>>>
>>>> Hi Sebastian,
>>>>
>>>> We will create a summary record of this discussion and post it.
>>>>
>>>> I do want to answer your questions but will keep it short here.
>>>> 1. We want single consistent, compact lightweight encap that works for v4, v6 and newer evolving formats.
>>> [SM] Which other L3 formats do you see evolving?
>>>
>>>> 2. Howsoever we want industry to migrate to v6, there is a pervasive usage of v4 and is there to stay. We must have a consistent solution for v4.
>>> [SM] There is the proposal for extension headers for IPv4, that would solve that issue, after all the environments you mention like DC are well controlled and requiring special IP or TCP options is not a show stopper... (as seen by CSIG itself which requires active buy in by network nodes and end points).
>> Hi Sebastian,
>>
>> I updated the draft on IPv4 extension headers
>> (https://datatracker.ietf.org/doc/draft-herbert-ipv4-eh/). I posted to
>> int-area list for discussion.
> [SM] Thanks a lot. Since this is a rather long draft it will take me a while to read through this ;) (but then this is far from my area of expertise so I likely will have have a few noob questions and no meaningful comments). Also I want to comb through the CSIG draft first (which is also on the long side, I wish more drafts were short and sweet, but I guess complex/extensive problems do not lend themselves to short and good enough solutions).
>
> Regards
> 	Sebastian
>
> P.S.: I do appreciate the goal of equalizing between the IP versions, instead of just hoping that IPv4 will go away any time soon.
>
>
<snip>

Please use the INTAREA mailing list to discuss this.

Please do not cross-post to tsvwg (but let us know if this is adopted).

Best wishes,

Gorry

(tsvwg co-chair)


_______________________________________________
ippm mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ippm