Re: [ppsp] AD review of draft-ietf-ppsp-peer-protocol-06

[Wesley Eddy <[e-mail]>]

On 5/10/2013 4:16 AM, Riccardo Petrocco wrote:
> Dear Wesley and All,
> 
>     Maybe someone can comment on whether alternatives like TFRC
>     were investigated or not?  TFRC is Standards Track, and intended
>     for providing a smoother rate like we talked about in the
>     past (LEDBAT may not be good for relatively rate-inelastic
>     flows), but I don't know if the feedback it requires would be
>     compatible with what the PPSPP framing provides or not.
> 
> 
> We are currently looking into alternatives to LEDBAT for the PPSPP.
> The majority of CCs will require some small modifications to the content
> of the DATA and ACK messages as defined in the draft.


That is good to hear.  I don't know if it's helpful or not, but
one example of how to implement TFRC in a non-DCCP UDP-based
protocol is:
http://tools.ietf.org/html/draft-eddy-tsvwg-saratoga-tfrc-03
I don't think it necessarily has to be as painful as some have
found in the past.


>     I also would not think it wise to change to something else
>     that doesn't match the implemented and deployed base, unless
>     the people implementing and deploying agree that something
>     else is clearly better and will be implemented and deployed
>     by them.
> 
> 
> We will also run some large experiments in a controlled environment to
> evaluate the effect and feasibility of the different solutions.


Excellent!


-- 
Wes Eddy
MTI Systems