Re: [multipathtcp] MPTCP Schedulers

[Olivier Bonaventure <[e-mail]>]

Phil,

> A couple of things personally I’d be interested in hearing about
> 
> -If there are differences for the schedulers deployed for end host 
> deployments of MPTCP and the proxy deployments

Yes. In the hybrid access networks use case, the schedulers are 
optimised to first use the DSL network and then overlfow over the LTE 
network when the DSL is full.

> -Scheduling algorithms that prefer one interface for policy reasons (eg 
> cost) ie the choice of interface is not primarily down to performance 
> differences between the interfaces (but presumably kicks in at some 
> point when it’s decided that the primary is down (how to decide that?) 

Detecting the failure of a primary path can be done by using existing 
TCP retransmission techniques. If the same packet needs to be 
retransmitted n times over a path, the stack can infer that this path is 
bad and should not be used for new packets.

> or that there isnt enough bandwidth on the primary interface (how to 
> decide that?)

The simple techniques in hybrid access networks is to simply measure the 
load of the DSL interface and take it into account when performing a 
scheduling decision


Olivier