headers
Sami Peltotalo | 9 Aug 2004 09:06
Picon
Picon

RE: [AVT] Commetns on draft-luby-avt-rtp-generic-fec-00.txt

Hi,

This is comment to Marcel's questions.

> > 4) How many FEC Schemes are planned in RMT? Is there a
> > recommendation, which
> > FEC scheme to use in which context?
>
> *** There is no planned limit, but new ones are continually popping up and
> being defined, I think there are three or four new proposals at
> the upcoming
> IETF, including a Reed-Solomon proposal (which needs some work
> still).  RFC
> 3452 defines the formal process of how to introduce new FEC Encoding IDs,
> and this process seems to be working quite successfully.
>
> > 5) Maybe a stupid question, but I was not able to find a
> concrete RMT FEC
> > scheme (except no code FEC scheme). Is there an implementation or a
> > specification of one FEC instantiation? I would like to see such spec to
> > understand how it is possible to write one specificatin which is
> > applicable
> > for RMT and AVT.
>
> *** THere are two types of schemes possible, fully-specified and
> under-specified.  There need to be a couple of more fully-specified, but
> there are quite a number of under-specified FEC schemes (most are
> proprietary) that have or will soon be registered with their own FEC
> Instance ID.  See RFC 3452 for how this works.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Marinho P Barcellos | 18 Aug 2004 11:12
Picon
Picon

Call for implementations

Hello

We work on a project that investigates the use of multicast in grid computing, 
and we would like to know if there are people in this list that have, or are 
aware of, Open Source implementations of reliable multicast protocols. For 
one-to-many file transfer, we know about:

* MDP, developed at the Naval Research Lab
* NORM implementation, also by the Naval Research Lab
* MCL library (NORM and ALC/FLUTE), developed at INRIA
* JRMS, by Sun Microsystems
* LGMP, by Markus Hofmann and others.
* FLUTE, developed at TUT
* TCP-XM, developed by Karl Jeacle

We know about DF too, but are actually looking for protocol implementations 
distributed under GPL or similar licenses. Any suggestions of other 
reasonably stable implementations?

Thanks in advance,
Marinho.

Dr. Marinho P. Barcellos
University of Manchester and BT Exact, UK
Senior Lecturer (on leave from Unisinos, Brazil).
Permalink | Reply | 0 comments |
headers
Sami Peltotalo | 26 Aug 2004 08:27
Picon
Picon

Simple XOR, Reed-Solomon, and Parity Check Matrix-based FEC Schemes

Hi all,

In our FEC scheme I-D we are referencing Reed-Slomon FEC instances in the
following way:

"1.2 Reed-Solomon FEC Schemes

   This document reserves two FEC Instance IDs, 0 and 1, for the
   Under-Specified FEC Encoding ID 129 name-space. Both of these FEC
   Instance IDs are for Reed-Solomon codes built on Vandermonde matrices
   [9]. A reference implementation for such codes can be obtained from
   the author of [9]."

Is this enough and right way to specify the FEC code?

Link to the I-D:
http://www.ietf.org/internet-drafts/draft-peltotalo-rmt-bb-fec-supp-xor-pcm-
rs-00.txt

Cheers, Sami Peltotalo
Permalink | Reply | 0 comments |
headers
Rod.Walsh | 27 Aug 2004 08:40
Picon

San Diego RMT minutes/proceedings

Hi Lorenzo and everyone

Any idea when the discussions and decisions of RMT San Diego will go public as minutes?

(Mike gave me the impression that much of significance was discussed, but those of us not in San Diego have
very little idea about it).

Apart from the "old" RMT (LCT,ALC,FLUTE,NORM,FECBB,*CC,...), the impact on the other items would be good
to understand rapidly (flute-iop, fec-supp-xor-pcm-rs, fec-supp-simple, avt-rtp-generic-fec).

Cheers, Rod.
Permalink | Reply | 0 comments |
headers
Rod.Walsh | 30 Aug 2004 14:37
Picon

RE: Simple XOR, Reed-Solomon, and Parity Check Matrix-based FEC Schemes

Hi Sami and RMTers

On a related issue, I have heard that it was decided in RMT/San Diego that "there should be only one encoding
ID per RFC, and thus per Internet Draft". (Note, FEC encoding ID, not Instance ID - apparanently it's OK to
have an unbounded number of the latter).

Would someone who was at the meeting please confirm or deny this rumour? 

The cut-off date for the IETF#60 proceedings input is 3.Sept so I am also looking forward to getting full RMT
commentry at http://www.ietf.org/proceedings/04aug/index.html by the end of the week.

Cheers, Rod.

-----Original Message-----
From: rmt-bounces <at> ietf.org [mailto:rmt-bounces <at> ietf.org]On Behalf Of ext
Sami Peltotalo
Sent: Thursday, August 26, 2004 9:28 AM
To: Rmt
Subject: [Rmt] Simple XOR, Reed-Solomon,and Parity Check Matrix-based
FEC Schemes

Hi all,

In our FEC scheme I-D we are referencing Reed-Slomon FEC instances in the
following way:

"1.2 Reed-Solomon FEC Schemes

   This document reserves two FEC Instance IDs, 0 and 1, for the
   Under-Specified FEC Encoding ID 129 name-space. Both of these FEC
(Continue reading)

Permalink | Reply | 501 comments |
headers
Lorenzo Vicisano | 30 Aug 2004 20:04
Picon
Favicon

Re: Simple XOR, Reed-Solomon, and Parity Check Matrix-based FEC Schemes

Rod,

the meeting minutes will be posted soon.

	thanks,
	Lorenzo

On Mon, Aug 30, 2004 at 03:37:35PM +0300, Rod.Walsh <at> nokia.com wrote:
> Hi Sami and RMTers
> 
> On a related issue, I have heard that it was decided in RMT/San Diego that "there should be only one encoding
ID per RFC, and thus per Internet Draft". (Note, FEC encoding ID, not Instance ID - apparanently it's OK to
have an unbounded number of the latter).
> 
> Would someone who was at the meeting please confirm or deny this rumour? 
> 
> The cut-off date for the IETF#60 proceedings input is 3.Sept so I am also looking forward to getting full
RMT commentry at http://www.ietf.org/proceedings/04aug/index.html by the end of the week.
> 
> Cheers, Rod.
> 
> 
> -----Original Message-----
> From: rmt-bounces <at> ietf.org [mailto:rmt-bounces <at> ietf.org]On Behalf Of ext
> Sami Peltotalo
> Sent: Thursday, August 26, 2004 9:28 AM
> To: Rmt
> Subject: [Rmt] Simple XOR, Reed-Solomon,and Parity Check Matrix-based
> FEC Schemes
>
(Continue reading)

Permalink | Reply | 501 comments |

Gmane