[Cake] Sce In cake testers wanted

Jonathan Morton chromatix99 at gmail.com
Sat Mar 23 19:53:29 EDT 2019


> On 23 Mar, 2019, at 10:28 pm, Kevin Darbyshire-Bryant <kevin at darbyshire-bryant.me.uk> wrote:
> 
> Sent a github PR to make it build - not run it yet.

LGTM, so I approved it.  I've reached the "bufferboat" and started to settle in, met Rod and Peteā€¦

I brought a Pi Zero W with me as a basic Linux test environment, but I need to find a mini to macro HDMI adapter to make it work with this TV.  Pete says he'll look for one to bring over tomorrow.

So now we probably have *two* working middlebox implementations, using different approaches, but we need a visualisation tool (hi Pete) and some sort of working TCP that uses it (whether using the sqrt rule or the 1/n rule, doesn't really matter - but preferably examples of both).  For the latter, we can temporarily get away with inferring stuff from standard tcptrace.  If there's a working implementation of AccECN for the feedback path, we might as well use that for demo purposes.

I want to be able to demonstrate SCE actually working as designed, in a single-queue implementation, to answer the main argument that some of the L4S folks have latched onto.  But I don't know how soon we'll be able to manage that.

 - Jonathan Morton



More information about the Cake mailing list