[Iccrg] ICCRG Meeting Presentation (TCP Libra and TCP Performance
cesar at cs.ucla.edu
Wed Aug 8 18:52:58 BST 2007
Thanks a lot for exposing our work.
Yes, I'm aware of the tool by by Gang Wang, Yong Xia, and David
Harrison. In fact, I have been investigating the other tool and also
reference in our website.
The difference should be that our tool aims comparative and
reproducible evaluations in complex network topologies and traffic
mix. So, as it is now, it may be considered in a later stage of TCP
evaluation, since, it focus on statistical behaviors. We are in touch
with the Yong Xia and Wang Gang and we look forward on a joint effort.
Thank again for your comments and we would appreciate further comments
On 8/7/07, Sally Floyd <sallyfloyd at mac.com> wrote:
> Cesar -
> > "TCP Evaluation Suite" - Assuming that the proposed congestion
> > control schemes would be in general used in the Internet, it is
> > important to study the interaction among flows of different schemes.
> > Thus, this new simulation "TCP evaluation suite" can assess
> > interaction among different congestion control proposals, by applying
> > the same experiment setup, including network configuration, flow
> > parameters, and workload. Finally, qualitative comparison automatic
> > graphs can be generated assessing flow-by-flow behavior of different
> > schemes.
> Thanks for the pointer to your TCP Evaluation Suite at
> "http://netlab.cs.ucla.edu/tcpsuite/". I have added a pointer to
> it on the TMRG web page at "http://www.icir.org/tmrg/".
> Gang Wang, Yong Xia, and David Harrison have also worked on a TCP
> evaluation suite, discussed in their internet-draft
> draft-irtf-tmrg-ns2-tcp-tool-00.txt on "An NS2 TCP Evaluation Tool".
> Their tool is available from "http://labs.nec.com.cn/tcpeval.htm".
> The two TCP evaluation suites have somewhat different roles and
> purposes; yours is based on NS-2 Linux, and is specifically focused
> on high bandwidth-delay environments. Still, it seems to me that
> it would be good for someone to look at the differences between the
> two evaluation suites, and see where it would be useful for each
> evaluation suite to borrow scenarios or approaches from the other.
> Take care,
> - Sally
More information about the Iccrg