[Iccrg] MulTFRC: still looking for feedback

[Michael Welzl]

"I repeat: ...

> -----Original Message-----
> From: dccp-bounces at ietf.org [mailto:dccp-bounces at ietf.org] On Behalf  
> Of
> Michael Welzl
> Sent: Monday, November 16, 2009 9:53 AM
> To: iccrg at cs.ucl.ac.uk list
> Cc: DCCP working group
> Subject: [dccp] Request for review of MulTFRC
>
> Dear congestion controllers!
>
> We are urgently looking for reviewers for MulTFRC, which is a
> new congestion control mechanism by Dragana Damjanovic
> and myself. A page with all documentation and code is here:
> http://heim.ifi.uio.no/~michawe/research/projects/multfrc/index.html
> I also made the usual ICCRG page, with links to some comments
> that we already received:
> http://tools.ietf.org/group/irtf/trac/wiki/ICCRG_multfrc
>
> Our intention is to get this approved as a WG item of the DCCP WG,
> with the plan to publish draft-welzl-multfrc-01.txt (the congestion
> control mechanism description) as an Experimental RFC, and
> also write a CCID specification for DCCP which should also become
> an Experimental RFC.
>
> I presented DCCP in the ICCRG meeting of the Stockholm IETF.
> Dragana presented it in the DCCP meeting in Hiroshima last week,
> where a part of the feedback was that the ICCRG should review the
> congestion control mechanism to be sure it's safe to deploy. The
> CCID specification (which doesn't yet exist) could then be discussed
> in the DCCP group only.
>
> MulTFRC is, as the name suggests, an extension of TFRC to
> support multiple flows (much like MulTCP is such an extension
> of TCP). To refine what the "safe to deploy" consideration could
> mean, I'd like to quote a statement from Michio Honda, from his
> comments about MulTFRC that he sent to the DCCP list:
> "I guess the IETF needs to establish the criterion and principle
> of N-TCP-friendliness before MulTFRC going ahead."
>
> PLEASE, volunteers, step up! We need to get reviews done
> before the next IETF meeting. In addition to in-depth reviews,
> general feedback is of course also extremely welcome, ideally
> to both this list and the DCCP list.
>
> Thanks!
>
> Cheers,
> Michael

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oakham.cs.ucl.ac.uk/pipermail/iccrg/attachments/20091208/21100e5d/attachment.html