Re: [Bloat] ECN & AQM Hall of Fame?

[Jim Gettys <jg@freedesktop.org>]

On 01/30/2011 08:03 PM, Dave Täht wrote:
>
>> On 01/30/2011 05:40 PM, Richard Scheffenegger wrote:
>>> Hi,
>>>
>>> wasn't there talk about getting a Hall of Fame for networks / operators, which are using and actively supporting AQM and ECN in their administrative domain?
>>>
>>> I just looked at some traces to troubleshoot SMTP with GMX (german freemail/hosting provider), and noticed that they are in fact negotiating for ECN on their mail servers:
>>>
>
> One data collection method I've been thinking about was using hooks on
> network chains (netfilter) to gather statistics on the actual incidence
> of ECN and SACK/DSACK.
>
> I have found a lot of places that negotiate ECN but have rarely seen ECN
> actually get used on the path.
>
> We'd need an ECN hall of fame and a separate SACK related one too.
>
Steve Bauer has been performing a very extensive survey of ECN on the 
Alexa top sites; he should have results in the next few weeks, for the 
CAIDA workshop.

Talk Title: A survey of the current state of ECN support in servers, 
clients, and routers

Talk Abstract: Explicit congestion notification (ECN) is a key building 
block in a number of ongoing standardization efforts [Conex] and 
research projects [Alizadeh]. This paper therefore sought to survey the 
current state of ECN support on the Internet, updating and extending a 
similar survey from 2008 [Langley]. (There are a number of reasons to 
suspect the state of ECN support may have changed including 'server-side 
ECN' becoming the default on recent versions of the Linux kernel.) In 
the process of conducting our survey we discovered that some routers 
incorrectly handle the ECN bits in the IP header, namely clearing the 
ECT bit. Given that this is a new impediment to using ECN, we sought to 
carefully measure exactly where this problem occurred. While measuring 
ECN support to web servers is straightforward, we developed novel 
active/passive hybrid approaches for collecting similar measurements of 
paths to clients. This is important because even if some servers support 
ECN, if the paths to clients contain impediments to ECN usage (which we 
show they do) the incremental deployment of ECN is harmed.

In short, ECN is no longer blocked much, but very few are actually 
turning it on.

News when it's available...
			- Jim