[Tmrg] sending rate of a TCP flow

[mallman at icir.org (Mark Allman)]

> 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?

With good modern TCP this hypothesis has always held OK for me.  E.g.,
see figure 6 in [1].  In older TCPs without SACK we often did find that
there was in fact quite a disparity between similar connections (but
digging up a written down reference to that experience is proving
difficult at the moment).  I don't think this "proves" the hypothesis,
but in offers some evidence.

allman


[1] Wesley Eddy, Shawn Ostermann, Mark Allman.  New Techniques for
    Making Transport Protocols Robust to Corruption-Based Loss. ACM
    Computer Communication Review, 34(5), October 2004.
    http://www.icir.org/mallman/papers/ceten-ccr-oct2004.ps



-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 193 bytes
Desc: not available
Url : http://mailman.ICSI.Berkeley.EDU/pipermail/tmrg-interest/attachments/20091026/9abec5ec/attachment.bin