mpls-linux-general

[mpls-linux-general] IPv6

[Luis F B. <ha...@in...>]

Hello all,

Is there anyone using IPv6 with this MPLS implementation?

My research team is doing some experiments with what we called "MPLS over
IPv6" by using IPv6 header fields (flow label, etc) to store MPLS label
information so that we can remove the MPLS labels and reduce the overhead,
fragmentation issues, etc. It's not a new idea. We are modifying NS-2 to
simulate that, but we'd like to see the real thing working, so we found
out that this implementation was available.

Please let me know if you had "good" experiences with IPv6 and this
MPLS implementation ;-)

Best Regards,
Luis F Balbinot
ha...@in...
Federal University of Rio Grande do Sul
Computer Science Department

Re: [mpls-linux-general] IPv6

[<j....@gm...>]

Am Samstag,  9. Juni 2001 19:36 schrieb Luis F Balbinot:
> Hello all,
>
> Is there anyone using IPv6 with this MPLS implementation?
>
> My research team is doing some experiments with what we called "MPLS over
> IPv6" by using IPv6 header fields (flow label, etc) to store MPLS label
> information so that we can remove the MPLS labels and reduce the overhead,
> fragmentation issues, etc. It's not a new idea. We are modifying NS-2 to
> simulate that, but we'd like to see the real thing working, so we found
> out that this implementation was available.
>

Sounds interesting, but how do you solve label stacking? This might be one of 
the biggest advantage of mpls.

best regards - Jörn

Re: [mpls-linux-general] IPv6

[Jing S. <js...@ca...>]

>

And, How the lable switching engine  and LIB built in linux?

regards

James Shen


>
> > My research team is doing some experiments with what we called "MPLS over
> > IPv6" by using IPv6 header fields (flow label, etc) to store MPLS label
> > information so that we can remove the MPLS labels and reduce the overhead,
> > fragmentation issues, etc. It's not a new idea. We are modifying NS-2 to
> > simulate that, but we'd like to see the real thing working, so we found
> > out that this implementation was available.
> >
>
> Sounds interesting, but how do you solve label stacking? This might be one of
> the biggest advantage of mpls.
>
> best regards - Jn
>
> _