On Wed, Aug 11, 2010 at 11:55 AM,  <[email protected]> wrote:
> In einer eMail vom 10.08.2010 23:54:23 Westeuropäische Sommerzeit schreibt
> [email protected]:
>
> For example, if multiple locator pairs all result in paths that coincide on
> a particular congested link, then the benefits of multi-pathing could be
> extremely limited.

Good question and this is hard to overcome in Internet.

>
> And: MPTCP's intelligent locator selection can never be a concerted action.
> Note, a congestion is not caused by one single IP-flow. Nor is it caused by
> a completely broken node (unless indirectly), which means that some flows
> should continue to pass the congested node while others should detour it.

One way to approach this problem could be that the transport protocol
is indicating to the network layer that the first subflow must always
take the shortest path through the network. The second and following
subflows belonging to the same session is market with a flag by the
transport protocol at the network layer stating "in case of congestion
(or close to congestion) on a forwarding link apply Valiant
Load-Balancing for these flows". This would achieve what Heiner is
pointing out - one sublfow  of the session is still taking the
shortest path and the other subflows are forwarded to explicit paths,
offloading the congestion point.

-- patte
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to