Re: [rohc] RE: Default decompression algorithms

Hi:

I will try to explain the scenario I am thinking of, and then try
to answer your questions.

Typically a proxy will be serving many users, let's say, thousands.
As such, when a terminal starts the SigComp procedures towards the
proxy, it might happen that the proxy has got a copy of the algorithm,
because the same algorithm was already uploaded by another terminal.

If this is the case, the decompressor at the proxy should announce that
those algorithms that are already uploaded, and therefore, don't require
to be uploaded once more.

If it happens that the terminal wants to use a different algorithm that
is not already in use at that proxy, then the terminal will know that
the algorithm is not uploaded and the terminal needs to uploaded.

You asked how many algorithms are going to be in the cache of a 
typical decompressor, and that question is related to how many users
a proxy can support simultaneously, and how many different algorithms.
In other words, it is implementation dependent, and the answer to the
question does not have an impact on the standardization in Rohc.

The cost of supplying the information: I don't have a strong opinion
on how this can be done, but I will refer to examples that are working
today in other protocols. For instance, RTP decided to use a profile
for Audio/Video codecs (RFC 1890). Known codecs are assigned a Payload
Type in RTP. So Payload Type 0 means PCM u-law codec, Payload 8 means 
PCM A-law codec. 

In SigComp, we could take the same approach, in a way that we define
a standard way to refer to known algorithms. The decompressor, in
the SigComp Capabilities Announcements would indicate the list of 
algorithms that are already uploaded (in a similar way of how SDP,
RFC 2327 negotiates codecs).

If there are other opinions how to fulfill the requirement, I would 
like to listen to them.

/Miguel

"Dr. Carsten Bormann" wrote:
> 
> > I also agree with the requirement whereby a decompressor informs
> > the compressor of the algorythms that are already uploaded, and
> > therefore, don't need to be uploaded again. This is just to
> > save bandwidth over the air interface and avoid unneeded uploads.
> > I think this is a very valid requirement.
> 
> How many algorithms are there going to be in the cache of a typical
> decompressor?
> What is the cost of supplying this information?
> Explain how this computes.
> 
> Gruesse, Carsten

-- 
Miguel-Angel Garcia                     Oy LM Ericsson AB
                                        Jorvas, Finland
mailto:Miguel.A.Garcia@ericsson.com     Phone:  +358 9 299 3553
mailto:Miguel.A.Garcia@piuha.net        Mobile: +358 40 5140002

_______________________________________________
Rohc mailing list
Rohc@ietf.org
https://www1.ietf.org/mailman/listinfo/rohc