Re: [tsvwg] path forward on L4S issue #16

[Ingemar Johansson S <[e-mail]>]

Hi

Thanks Jake, this clarifies things in this respect. 
I believe/hope that larger content providers and ISPs can be easier to address with operator guidelines that Wesley (David?) suggested, and that can perhaps work as we see a reasonably good industry support behind L4S. And after all a decision to bleach ECT(1) need to be preceded by some kind of investigation about the options ?
Individuals that install custom CPEs are a more uncertain element of course. We discussed this a bit in this thread, there are possibilities as well as challenges here.

With that said, I believe that it is now fair that I do a multiplicative decrease and give this group a chance to breathe 😊

/Ingemar

> -----Original Message-----
> From: Holland, Jake <[e-mail]>
> Sent: den 17 juni 2020 23:12
> To: Ingemar Johansson S
> <ingemar.s.johansson=[e-mail]>; Rodney W. Grimes
> <[e-mail]>; Jonathan Morton <[e-mail]>;
> Sebastian Moeller <[e-mail]>
> Cc: Ingemar Johansson S <[e-mail]>;
> [e-mail]
> Subject: Re: [tsvwg] path forward on L4S issue #16
> 
> Hi Ingemar,
> 
> One clarification:
> 
> On 6/17/20, 9:27 AM, "Ingemar Johansson S"
> <ingemar.s.johansson=[e-mail]> wrote:
> > So far I see only one firm evidence that RFC3168 AQMs are deployed and
> > enabled and these are related to the OpenWrt project, manifested in a
> > few upgradeable devices (and some not upgradeable) that can be bought
> > in retail stores. The rest is unconfirmed.
> 
> If you're referring to the observations I sent, they are not linked to
> OpenWRT.  I was looking at Akamai production traffic that saw ECE marks
> while delivering content to end users.
> 
> Although many of the marking devices were likely CPE devices deployed by
> individual end users, there's reason to believe that several ISPs (what looks
> like a growing number) have some kind of managed deployment that has a
> much greater prevalence within their ASN than can be easily explained by
> individual users installing a custom CPE.
> 
> I don't have any information on whether the marking devices inside these
> wider deployments within specific ASNs are predominantly CPE devices or
> whether they're one of the many routers or access devices that support CE-
> marking that they have now chosen to enable, after historically running it
> disabled.  (This is not implausible, because the amount of traffic that will
> respond to the marks has changed substantially in the last few years.)
> 
> HTH.
> 
> Best regards,
> Jake
>