Re: [Congress] CONGRESS is about ready to go

[Matt Mathis <[e-mail]>]

Both of the items I mentioned cross layers.    I suggest putting them in
the charter as a defense against people elsewhere in the IETF objecting to
us meddling in their affairs.

On Thu, Feb 23, 2023 at 8:59 AM Bob Briscoe <[e-mail]> wrote:

> Matt,
>
> On 21/02/2023 22:15, Matt Mathis wrote:
>
> Looks great!   I would like to suggest some more some "in charter" items:
>
>    -  Interactions between congestion control and lower layer algorithms
>    that alter packet timing or thin cumulative ACKs.   Such algorithms are
>    pervasive in half duplex link layers, such as WiFi, where channel
>    arbitration overhead often dominates performance in terms of both latency
>    and bandwidth.
>
>
> [BB] I'd agree that this is a good pair of generalizations to be addressed
> by the WG. Do you envisage writing guidelines for implementers of the lower
> layer algos to be in scope, or only guidelines for implementers of
> end-system algos? My vote would be for both.
>
I think the first step would be a document to better define the problem and
some of the tradeoffs, trying as much as possible to be agnostic about any
particular treatment.  Unfortunately there are strong local optima that are
globally non-optimal.


>
>    -  Upper layer or application algorithms that implicitly or explicitly
>    defeat transport layer algorithms intended to protect the network.  An
>    example would be an application that aborts and restarts transport
>    connections on a fixed interval timer, implicitly defeating the transport
>    layer's exponential RTO backoff.
>
>
> [BB] Were you prompted to include this by some current implementation
> attempting this?
>
This class of bugs is pervasive.  For example they are likely to be present
in any application that does fixed timer based requests (or retries),
without regards to the completion or error statuses of prior requests.  A
few cases can be quite harmful, such as a streaming video player that
requests new chunks while prior chunks are still in flight or stalled.

Most of the time these bugs don't cause problems, but buggy code deployed
at scale can be very disruptive.  All examples that I can think of stem
from inappropriate actions on errors, where the application responds to
errors either by increasing the presented load or wasting capacity.   For
example, if git-clone fails due to a network error, you have to rerun the
entire operation, wasting the data already delivered.

The point being that CC principles really need to be applied to the entire
stack, not just the transport layer.

I don't think it belongs in a list of things the WG is chartered to work on
> though (a charter isn't normally a place where you would find a list of
> things to /not/ do). I don't think it would even need to be in the ICCRG
> charter - we don't need any (more) research to prove that this would cause
> congestion collapse. It is surely just sthg that the proposed RFC5330bis
> would say is highly dangerous, explain why, and say "thou SHALT NOT".
>

I have met too many application designers who think that transport (TCP) is
fully capable of protecting the network from poorly behaving applications.
We need some tools to counteract this mindset.


>
>
> Bob
>
>
> Thanks,
> --MM--
> Evil is defined by mortals who think they know "The Truth" and use force
> to apply it to others.
>
>
> On Tue, Feb 21, 2023 at 10:19 AM Jeff Tantsura <[e-mail]>
> wrote:
>
>> Hi Martin,
>>
>> From my point of view, the charter looks good and is ready to be
>> submitted to the IESG.
>> Thanks for the good work, looking forward to working together!
>>
>> Thanks,
>> Jeff
>>
>> On Feb 21, 2023, at 9:12 AM, Martin Duke <[e-mail]> wrote:
>>
>> Hello transport enthusiasts,
>>
>> I'm pleased to announce that all the pieces are in place to move forward
>> with congress, the working group focused on congestion control and related
>> topics.
>>
>> (1) If you haven't already done so, and are interested, please subscribe
>> to the congress mailing list:
>> https://www.ietf.org/mailman/listinfo/congress. This is the last time
>> I'll crosspost (Bcc:) to multiple WG lists, and I encourage others to stop
>> as well.
>>
>> (2)  Although the proposed charter has no formal standing at this time,
>> I'm initiating a "last call". Please have a look and file issues by the end
>> of this week:
>> https://github.com/martinduke/congestion-control-charter
>> The only recent additions to the scope are Fair Queueing and rate pacing.
>> I'd like to hand this to Zahed, our sponsoring AD, to run through the IESG
>> process next week.
>> --
>> Congress mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/congress
>>
>>
>> --
>> Congress mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/congress
>>
>
>
> --
> ________________________________________________________________
> Bob Briscoe                               http://bobbriscoe.net/
>
>

-- 
Thanks,
--MM--
Evil is defined by mortals who think they know "The Truth" and use force to
apply it to others.