[Ecn-sane] [Bloat] [tsvwg] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104

Jonathan Morton chromatix99 at gmail.com
Wed Mar 20 19:03:32 EDT 2019


> On 21 Mar, 2019, at 12:56 am, Sebastian Moeller <moeller0 at gmx.de> wrote:
> 
> What they should have done, IMHO is teach their AQM something like SCE so it can easily react to CE and drops in a standard compliant TCP-friendly fashion, and only do the clever window/rate adjustments if the AQM signals ECT(1)…

It's not even hard to teach DCTCP to respect SCE, as far as I can tell.  All you have to do is start with a clean copy of NewReno with ABC, then subtract half of all SCE-marked bytes from the cwnd.  That really is it.

 - Jonathan Morton



More information about the Ecn-sane mailing list