Hi Aaron,
I understand your argument more as an argument for equitable sharing/FQ than
for AQM, even though both complement each other quite well.
The biggest point for FQ in my opinion is, while this is certainly not the
optimal capacity share regime for all traffic mixes, but without an oracle at
the bottleneck (or robust un-gameable information about relative priority
encoded in the packets themselves) it is the one sharing regime that will never
be pessimal, and should always be good enough and also per definitionem
compatible with net neutrality rules, since no flow gets an advantage over
another one. That said FQ does not solve all pathologies and is for example
gameable by splitting a transfer into many concurrent flows, but single queue
AQMs and even a dumb FIFO behave similarly the same pathology already, so FQ
does not make matters worse.
Regards
Sebastian
> On May 29, 2021, at 00:28, Aaron Wood <[email protected]> wrote:
>
> I think one of the big advantages that AQM has is that it doesn't know, or
> care, who the flow is. It can't, itself, violate NN concerns because it has
> no knowledge with which to do so.
>
> Instead, it punishes the greed flows that try to use more than their fair
> share of the available bandwidth. It doesn't really favor one protocol, or
> provider, or site, or anything else, because it's not aware of them, as such.
> Instead it just stops the kid that's trying to take all the candy from the
> candy bowl, to make sure there's enough for everyone else in line.
>
> The working could perhaps be softened from "punished", perhaps, depending on
> the audience.
>
> "Everyone gets an equal slice, up to their fill" is another way of looking at
> it, I think.
>
> On Wed, May 26, 2021 at 11:36 AM Dave Taht <[email protected]> wrote:
> oops, wrong link for the that plea to both sides of the debate. The
> correct link was:
>
> http://blog.cerowrt.org/post/net_neutrality_customers/
>
> On Wed, May 26, 2021 at 11:32 AM Dave Taht <[email protected]> wrote:
> >
> > I made the most articulate pleas I could to both sides of the debate on
> > this:
> >
> > https://blog.apnic.net/2020/01/22/bufferbloat-may-be-solved-but-its-not-over-yet/
> >
> > I do think that now that the scandal here has reached a peak:
> > https://ag.ny.gov/sites/default/files/oag-fakecommentsreport.pdf
> >
> > that perhaps an honest appraisal of both AQM and fq+aqm technologies can be
> > had
> > in public, again. I'd be willing to reach across the isle on this, and
> > patiently explain stuff to lawmakers where their thinking is
> > incorrect, in order to finally fix bufferbloat.
>
>
>
> --
> Latest Podcast:
> https://www.linkedin.com/feed/update/urn:li:activity:6791014284936785920/
>
> Dave Täht CTO, TekLibre, LLC
> _______________________________________________
> Bloat mailing list
> [email protected]
> https://lists.bufferbloat.net/listinfo/bloat
> _______________________________________________
> Bloat mailing list
> [email protected]
> https://lists.bufferbloat.net/listinfo/bloat
_______________________________________________
Bloat mailing list
[email protected]
https://lists.bufferbloat.net/listinfo/bloat
