Re: [rmcat] Review of draft-ietf-rmcat-scream-cc-01

[Karen Elisabeth Egede Nielsen <[e-mail]>]

HI,
>>>
>> [Karen ] Yes I think that can be a good idea, but I am not sure that
>> the above mapping of terms is right  - or is it  ?.
>> I very likely can be mistaken, but with screams multiple stream view
>> then I have thought that the following was the case:
>>
>> Scream Transmission scheduler is in fact a true transmission scheduler
>> as it takes packets from the multiple RTP queues and schedule the
>> packets (following whatever multiple-stream scheduling policy/stream
>> priority
>> relevant)
>> for consummation of the global sending rate/the congestion window
>> calculated and thus transmit the packets on the UDP socket (after
>> which they eventually will go to the NIC for transmission). I further
>> had assumed that pacing (vis-à-vis cwnv) in this design is implemented
>> as part of the transmission scheduler and not as part of the RTP queue
>> logic.
>> This model at least corresponds to SCTP's model for scheduling of the
>> multiple streams for consummation of CWND and transmission.
>
>If I have red the current draft correctly, everything you've described here
>was
>moved to the appendix and all that is left in the main part is the
>calculation of
>the actual sending rate. In this case it would map (to NADA). Probably you
>want one image in the main part that is similar to the image in NADA and
>then
>have another image in the appendix where you show multiple sources and an
>actually transmission scheduler.
>
[Karen ] Yes - I see your point. One note though is that in a multi-stream
model then pacing
logically is instantiated by the multi-stream transmission scheduler which
consumes the packets from the RTP queues
and not logically instantiated in the individual RTP queue objects.
I think that it is important that the terminology (and the definition of the
objects within the framework one can say)
accommodates for a both a single RTP stream and a multi-stream solution - Or
?

BR, Karen

>Mirja
>
>
>>
>> BR, Karen
>>
>>> --
>>> Zahed
>>>
>>> ==================================================
>>> ANM ZAHEDUZZAMAN SARKER
>>>
>>>
>>> Ericsson AB
>>> Services, Media and Network Features
>>> Laboratoriegränd 11
>>> 97128 Luleå, Sweden
>>> Phone [phone]
>>> Fax [phone]
>>> SMS/MMS [phone]
>>> [e-mail]
>>> www.ericsson.com
>>>
>>> ==================================================