[Iccrg] sending rate of a TCP flow

[Douglas Leith]

Depends on the buffering, but here's one reference that applies to  
drop-tail queues (fluid work is all for AQMs):

A positive systems model of TCP-like congestion control: Asymptotic  
results. IEEE/ACM Transactions on Networking, 14(3), pp616-629, link: http://www.hamilton.ie/net/unsynchronised_final.pdf

Might also be worth noting that the TCP-friendly equation (I presume  
you mean Padhye's fluid model ?) is really for a single flow in a  
"bath of noise" i.e. where packet loss is independent of TCP  
congestion control action.  This is definitely *not* valid for  
multiple flows sharing a single queue in a dumbell topology since the  
loss rate experienced by each flow can (and generally will) be  
different if they have different RTTs - its easy to see this by  
considering an example where flow backoffs are synchronised.

Doug

www.hamilton.ie

On 23 Oct 2009, at 04:42, aydin at mail.eecis.udel.edu wrote:

> Hello,
>
> Based on the TCP-friendly equation, sending rate of a TCP flow  
> depends on
> RTT, MSS, and loss rate.
>
> hypothesis: When N long-lived TCP flows (with the same RTT and MSS  
> values)
> share a bottleneck link in a dumbbell  topology (all the edge links  
> have
> the same delay and bandwidth), I expect each TCP flow to have the same
> sending rate -- Assume no other cross traffic in the network and  
> drop-tail
> queues.
>
> Do you know any references (simulation-, emulation-, or real
> experiment-based studies) that prove this hypothesis or the opposite?
>
>
> thanks,
> ilknur Aydin
> ps. Please let me know if there is a more appropriate mailing list  
> to post
> this question.
>
>
> _______________________________________________
> Iccrg mailing list
> Iccrg at cs.ucl.ac.uk
> http://oakham.cs.ucl.ac.uk/mailman/listinfo/iccrg