Strange Attractors in Network Flows

[Tim Bass <[e-mail]>]

Hi.

The topic is dissipative systems with a degree of 'randomness' similar
to Shaw's observation of strange attractors in the 'leaky faucet model'
in the tales from the 'crypts of chaos'.

I was thinking about the model and it appeared to me (conjecture) that
IP flows in large networks could be modelled as a similar dissapative
system with random packet interarrival times; dissapative (of course)
because packets are dropped and this is 'more than likely'  
a nonlinear dynamic process that could be modelled.  

Hence, based on this conjecture, it seems that chaos theory is quite
applicable to IP internetwork flows and therefore is useful to
developing models of large internets...... BUT...

As usual, there are few unique ideas and applications, so I believe
that *someone out there* has been working in this area and has published
some IP CHAOS THEORY papers.  I have not, however, seen any such
publications referenced.   Does anyone on big-internet (Noel?)
know of any publications in this area?  If so, could they forward
or post the papers ( I have plenty on chaos theory, just nothing
on IP-CHAOS).

If not, is anyone interested on big-internet in working on looking
for strange attractors in chaotic large IP internetwork flows?

Is there a discussion group for this?  {Big-internet might be
a good place to introduce a new topic, based on the minimal
technical traffic lately.} Or, is application of chaos theory
to large IP networks inappropriate for the big-internet list?

(private e-mail also appreciated)

Kind Regards,

Tim

-- 
We're just two lost souls 
	swimming in a fish bowl, 
		year after year,

Running over the same old ground. 
	What have we found? 
		The same old fears.

	Wish you were here.
		-Roger Waters