RE: [PWE3] PWE3 new WG items?

[benjamin.niven-jenkins at bt.com]

> I  think that  all  you're really  saying is  that  the QoS  
> belongs to  the
> transport (the  enclosing LSP or  the L2TPv3 session),  not 
> to the  PW.  (Of
> course, it's hard to see through the gibberish of buzz words.) 
> 
> The opposite point of  view is that you need to know the  PW 
> QoS in order to
> select a transport  LSP or L2TPv3 session properly,  one 
> endpoint might want
> to  signal this  to the  other, and  the  PW QoS  might even  
> impact (or  be
> impacted by) some  characteristic of the AC.  I think  this 
> is what Himanshu
> was saying, and prima facie it seems right to me.  

I agree that you need to know the QoS a PW requires in order to associate it with a transport LSP etc.  However as PW signalling is basically piggybacked onto LSP signalling what is the motivation for a PW specific QoS signalling mechanism (as suggested by Himanshu) when LSP QoS signalling mechanisms already exist?


Ben

_______________________________________________
pwe3 mailing list
pwe3@ietf.org
https://www1.ietf.org/mailman/listinfo/pwe3