Re: [tsvwg] Draft TSVWG minutes from the meetings in Montreal

[Dave Taht <[e-mail]>]

Jonathan Morton <[e-mail]> writes:

>> On 10 Aug, 2019, at 10:46 am, Gorry Fairhurst <[e-mail]> wrote:
>> 
>> The Chairs have uploaded draft notes from the meetings to the materials.
>> 
>> https://datatracker.ietf.org/meeting/105/materials/minutes-105-tsvwg-03
>> 
>> Please let us know if there are any factual errors in the notes or corrections
>> to names, etc. Thanks again to the volunteers who helped take these notes!
>
> I have a few proposed corrections:
>
>> Jonathan Morton: I have 4 topics:
>> f) IETF approval is required to use ECN.  
>> g) Incremental deployment requires compatibility, which has not been demonstrated so far.
>> a) Effective congestion control is required. FQ_Codel is the most deployed
>> AQM. With the current Codel, the time to reach the correct marked rate would
>> be 4 seconds (?). 4 seconds for a link with 10ms is not effective.
>> b) Fair queueing is fairly robust. My concern is when there are two
>> consecutive bottlenecks. In our experiments we have found issues.
>
> Referring to the notes I was reading from:
>
> Under point F, I was not referring to ECN use in general - that would be absurd
> as it's a current RFC - but to the Congestion Response/Marking Differences
> experiments provided for in RFC-8311, due to the effective redefinition of the
> CE codepoint proposed by L4S.
>
> Under point A, I was referring to Codel in general, not FQ-Codel, although most
> actual deployments of Codel are indeed as part of FQ-Codel.  The inadequate
> response of DCTCP (and thus presumably Prague) to Codel is of only slightly less
> concern in FQ form, due to the consecutive-bottleneck problem alluded to in
> point B (where the upstream bottleneck may be a dumb FIFO).


> The following would better capture the above:
>
> f) IETF approval is required for ECN semantic changes, even under RFC-8311.
> [...]
> a) Effective congestion control is required. Codel is the most deployed AQM.  (etc.)
>
>> Wes: Does anyone want to bring up IPR?
>> Rod: There are people who want to speak to IPR are not present.
>> Jonathan: Dave Taht wants to speak to this later, he could not join.
>> Chairs: Please read the IPR summary and decide on your own position.
>
> Here, the name I mentioned was Toke Hoiland-Jorgensen, not Dave Taht.  I'm sure
> Dave also has much to say on the subject, but it was Toke who had specifically
> said he wanted to comment that day.

I've said my piece here, actually. It's in the hands of the lawyers now,
fronted by toke.

>
>  - Jonathan Morton