[e2e] Are we doing sliding window in the Internet?
David P. Reed
dpreed at reed.com
Mon Jan 7 07:22:29 PST 2008
Lars - I'm always interested in learning new things. Such an attempt
would be fruitful - I have been wrong in the past, and can admit I am
wrong. Sorry if my expression of general frustration with the
sociology of the IETF and research communities was too "hot-headed". If
I had stated it in less spicy terms, there would still be a strong
argument that others could choose to rebut.
It was not triggered specifically by any one person's views - more an
integration of a set of views that have been bugging me for years.
Which is why I did not include any quoted text.
Perhaps my spiciness discourages discussion. I would hope it would do
the opposite, perhaps encouraging people to marshal real arguments, real
data to disprove what I argue for.
Lars Eggert wrote:
> Hi, David,
> since your message has me in the "To" field, I'm guessing it was
> triggered by one of my earlier emails.
> If that's the case, either I'm not expressing my points clearly
> enough, or you're significantly over-interpreting them. Or both.
> Usually, I'd try to clarify things at this point, but reading the rest
> of your email somehow doesn't leave me with the impression that such
> an attempt would be fruitful.
> On 2008-1-7, at 16:46, ext David P. Reed wrote:
>> 99% of the effect of the original TCP congestion control "innovation"
>> was due to a simple and robust principle: when a packet is lost, cut
>> the traffic being sent *drastically*. This converted a tendency of
>> the community to turn packet losses into a positive feedback loop of
>> retransmission into a shared understanding that "positive feedback
>> loops are bad".
>> That was a community-scale learning that was useful.
>> I would argue that the idea that TCP is now "near perfect" as a
>> standard that *must be enforced lest the Internet enter congestion
>> collapse* has almost no real-world evidence.
>> In fact, very little of the traffic on the network today comes in the
>> form of end-to-end unbounded-rate-demanding streams. TCP's
>> congestion control (and the definition of "TCP compatible" fairness,
>> too!) is "tested" only for such a trivial case. No One At All tests
>> TCP congestion control in a world of HTTP, VoIP, streaming video from
>> YouTube, Skype, etc. The "gurus" of the IETF that study protocol
>> *documents* have intuitions that are seriously flawed, if only
>> because they don't actually have any data about the traffic ANYWHERE
>> in the damn network. (kc claffy would die to have such data).
>> So it seems that a bunch of self-elected high-priests of IETF have
>> taken it upon themselves to claim God-like, experiment-free knowledge
>> of the Internet as it really is, and to use that knowledge to claim
>> that TCP is the Gold Standard that must be obeyed lest the network
>> finally do what Bob Metcalfe erroneously said it would do and ate
>> whatever piece of clothing he ate as a result.
>> Do science and real engineering, please. Yeah, the AIMD algorithm
>> in TCP saved the Internet's ass. But it is not Holy Writ.
More information about the end2end-interest