In einer eMail vom 08.08.2010 19:13:29 Westeuropäische Sommerzeit schreibt  
[email protected]:

c)  should this multipathing feature belong to the network layer, i.e.
creating  e.g. a generic congestion mechanism for all transport
protocols or is  multipathing something that the transport protocol
layer should take care  of?



Excellent question. Maybe we should have a vote and count how many believe  
that multipath is a network layer issue and how many believe it is a  
transport layer issue.
 
IMO, it is  definitively a network layer issue, and also that MPTCP  and 
CONEX are misguided concepts and activities. Furthermore:
 
It seems to be a long IETF-tradition to develop philosphies based on  
"knowing nothing about the network":
 
-  DiffServ: all based on just the router's own waiting  queues for 
incoming packets
- MPTCP, SHIM6, multihoming as is: all without hope that the network layer  
can ever become smart enough to solve the addressed problems (e.g.resolving 
 congestions not by blind  try-and-error alternative paths but by well  
concerted selection of detours).
 
Why must this be so forever ?
 
Heiner
 
 
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to