Peter,
Why do you use the flow label for QOS parameters, rather than
the traffic class octet that was defined for that purpose in
RFC 2474?
Brian
Peter Sevenich wrote:
>
> Hi,
> we have two years of experience in using the IPv6-flowlabel to transmit
> timecritical data (for example vocoded voice) and conventional data at the same
> time over heterogenouse networks including those of low bandwhith ( in the
> moment ISDN, GSM, HF, seriel modem, gateway to usual phone etc..).
> We use the flowlabel to transmit flow_id and QoS-parameters for channel resorce
> management in any packet. In the moment we have implementationes on Linux
> Solaris and BSD and start inernational tests.
> We are interesting in any work around the flowlabel. With RFC 2553 we are
> unlucky, because it is said that the system sets the (random) value of
> flowlabel. For us it is important that the aplication can set and read this
> value. I think a good solution for the flowlabel should include this
> possibility.
>
> We are interesting in any colaboration and discussion to write a draft.
>
> Best Regards
> Peter
>
> Thomas Eklund wrote
> >
> >We have an idea of what to use it for.
> >I'm currently writing a draft about "traffic engineering in ipv6" like I said
> to you Bob.
> >A draft proposal is finished quite soon and the intention is to see if there is
> some interest to use the flowlabel for traffic engineering and If you would like
> to co-author it thats perfect...
> >
> >Anyway I have asked for a slot on the IPng meeting for a couple of weeks ago
> and I hope that we can have a good discussion about this at the next IETf
> meeting
> >
> ><Thinking about the discussion on what to do with the IPv6 flow label field,
> ><I have couple of thoughts.
> ><
> ><The discussion about the best working group for this work is somewhat
> ><academic until there is a concrete proposal (i.e, an internet draft) for
> ><what do with it.
> >
> >You will have it before the next meeting as we discussed.
> >
> >Best Regards
> > Thomas Eklund
> ><The intent of the words in the draft charter was to say that if someone has
> ><a concrete proposal for how to use the IPv6 flow label field, they should
> ><present it to the IPng w.g. It will be easy once that happens to move it
> ><to a more appropriate w.g. if it makes sense.
> ><
> ><Just because we have the IPv6 flow label bits "available" it is not at all
> ><clear we want to do anything with them today. Our experience with IPv4 is
> ><that it would have been very nice to have had some open bits available for
> ><future use. The IP world will change in unexpected ways. These bits may
> ><become very useful 5-10 years from now. They may be needed for something
> ><that has not been thought of today.
> ><
> ><It might be much better to encourage experimentation with the flow label
> ><bits than to standardize something now. Lets use them wisely.
> ><
> >< Bob
>
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> + Peter Sevenich +
> + FGAN/FKIE/KOM +
> + Neuenahrer Str. 20 +
> + D-53343 Wachtberg-Werthhoven +
> + Germany +
> + Phone +49-228-9435-317 +
> + Fax +49-228-9435-685 +
> + e-mail [EMAIL PROTECTED] +
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
> --------------------------------------------------------------------
> IETF IPng Working Group Mailing List
> IPng Home Page: http://playground.sun.com/ipng
> FTP archive: ftp://playground.sun.com/pub/ipng
> Direct all administrative requests to [EMAIL PROTECTED]
> --------------------------------------------------------------------
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
