Re: [rmcat] WG last call: draft-ietf-rmcat-video-traffic-model-04

[Colin Perkins <[e-mail]>]

Thanks - this all looks good.
Colin



> On 19 Jul 2018, at 18:50, Xiaoqing Zhu (xiaoqzhu) <xiaoqzhu=[e-mail]> wrote:
> 
> Thanks, Colin, for your input. We have updated the draft to version -05 accordingly.Please see detailed responses below inline (tagged [XZ]).
> 
> Best,
> Xiaoqing
> 
> From: Colin Perkins <[e-mail] <mailto:[e-mail]>>
> Sent: Sunday, July 15, 2018 2:54 PM
> To: Fu Jiantao
> Cc: rmcat WG; [e-mail] <mailto:[e-mail]>
> Subject: Re: [rmcat] WG last call: draft-ietf-rmcat-video-traffic-model-04
>  
> Thanks for the comments!
> 
> After reading it again, I’d also note that:
> 
> The draft talks about “RMCAT algorithms”, but should probably talk about “RTP congestion control algorithms”, since the final RFC will outlive the RMCAT working group.
> 
> [XZ]  Good point. I've replaced the mention of RMCAT in the main text to "RTP congestion control algorithms.".   Not sure what to do with several of the references though.  For our own github code on "Syncodec" we can change the title later.  For the pointers to previous interim presentations: can we leave it as is? 
> 
> [XZ]  These questions can probably be deferred till the draft is in shepherd mode? 
> 
> The draft needs a security considerations section. I can’t of anything to include other than a statement about the importance of congestion control to stable operation of the network, and for RTP flows the importance of testing congestion control with realistic traffic.
> 
> [XZ] Thanks for the suggestion. I've added on section with the wording below: 
> 
>  It is important to evaluate RTP-based congestion control schemes
>    using realistic traffic patterns, so as to ensure stable operations
>    of the network.  Therefore, it is RECOMMENDED that candidate RTP-
>    based congestion control algorithms be tested using the video traffic
>    models presented in this draft before wide deployment over the
>    Internet.​
> 
> 
> Cheers,
> Colin
> 
> 
> 
> 
>> On 12 Jul 2018, at 21:59, Fu Jiantao <[e-mail] <mailto:[e-mail]>> wrote:
>> 
>> Hi, 
>> 
>>   Some comments below:
>> 
>>   1. In Abstract 
>> 
>> "
>>    The second model is trace-
>>    driven, and emulates the encoder output based on actual encoded video
>>    frame sizes from a high-resolution test sequence.
>> "
>> 
>> is that based on both encoded video frame interval and frame size? 
>> 
>> 2. In section 4, in the graph and description is misleading to me, the synthetic video codec takes raw video frames, but not in realtime like real video codec,  it's two steps process, the first step is to 
>> 
>>  1) for statistic model, fitting from emperical data to get parameters
>> 
>>  2) for trace-based model, gather meta data like frame size, frame type etc 
>> 
>> The first step is like pre-process, and the second step is to feed that into the synthetic video codec in realtime.
>> 
>> 3. In section 4. "Frame resolution XY", that is actually the target frame resolution, right? this may also be affected by subscription from remote receiver, for example when remote is a mobile client with a small screen.
>> 
>> - Jeromy
>> 
>> Colin Perkins <[e-mail] <mailto:[e-mail]>> 于2018年6月20日周三 上午1:49写道：
>> This is to announce a working group last call on “Modeling Video Traffic Sources for RMCAT Evaluations” (draft-ietf-rmcat-video-traffic-model-04). 
>> 
>> Please send any final comments to the working group mailing list and the authors by 20 July 2018 (the date of the RMCAT session at IETF 102). If no substantive comments are received by that time, we intend to submit this draft to the IESG for publication as an Informational RFC.
>> 
>> Colin (for the RMCAT chairs)
>> 
>> 
>> 
>> -- 
>> Colin Perkins
>> https://csperkins.org/ <https://csperkins.org/>
>> 
> 
> 
> 
> -- 
> Colin Perkins
> https://csperkins.org/ <https://csperkins.org/>
> 
> 
> 
> 
> 



-- 
Colin Perkins
https://csperkins.org/