[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [vnrg] FlowVisor : A Network Virtualization Layer



On Wed, Nov 10, 2010 at 7:16 PM, Martin Stiemerling
<Martin.Stiemerling at neclab.eu> wrote:
> Rob,
>
> You say:
>
>> - network topologies that are independent of the physical topology
>> (currently FlowVisor is restricted to subsets of the physical
>> topology)
>
> Does this also exclude right now the aggregation of multiple links to one virtual link?

Just so I understand you, you mean, for example, something like
channel bonding, where packets from one virtual link are load-balanced
across multiple physical links?

Right now, FlowVisor itself does not support this.  So, there are two
qualifications there: 1) this is just a limitation of the current
implementation and would be fairly trivial to add, and 2) one reason
why it's not currently implemented is that support for this is still
being added to the OpenFlow protocol (see the proposed OpenFlow 1.1
extensions off of openflow.org), so once that's in place, it will be
much cleaner to implement, architecturally speaking.   That, and no
one has asked for it :-)

I apologies, but it looks like I will be unable to join the
presentation (virtually or otherwise).  But I do appreciate the
conversation that has been taking place on the list and look forward
to continuing these dialogs.

- Rob
.

Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.