[Iccrg] A review of the CUBIC draft

[Kashif Munir]

Hi all,

 

I have read the CUBIC draft and have seen the extensive simulations and
experimental results related to CUBIC on the authors' pages. Here are my
comments and suggestions about it.

 

The language of the CUBIC draft is simple and clear. The draft almost
completely follows the guidelines for specifying the new congestion control
algorithms [RFC 5033]. The CUBIC protocol can be considered safe for
deployment as it does not perform worse than TCP in difficult environments
like wireless networks. I believe it should be published as an experimental
RFC. 

 

I have a few suggestions that may possibly improve the quality of the draft.

 

-   The draft must include an explanation on why the value of C (the CUBIC
parameter) is taken equal to 0.4. The CUBIC paper does not discuss it
either.

 

-   Regarding RTT-fairness among CUBIC flows, the explanation must be clear
and consistent throughout the document. It must be clearly specified that
the CUBIC flows are only RTT-fair when they have same RTT and by taking some
time for convergence to fair share otherwise the CUBIC flows have their
bandwidth shares linearly proportional to the inverse of their RTT ratio.

 

-   The authors must explain in the draft about the assumption, "we believe
that under wired Internet, use of the linear share notion seems more
reasonable than equal share or a higher order shares".

 

 

 

Best Regards

 

Kashif Munir

-------------------

PhD Candidate
Institute of Computer Science
University of Innsbruck, Austria
Homepage: http://dps.uibk.ac.at/~kashif/


 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oakham.cs.ucl.ac.uk/pipermail/iccrg/attachments/20080526/0d61e305/attachment.html