Hey Rob, Thanks for the slides, this is a great addition to the existing OpenFlow literature. Any chance you could also tell us, where this was presented, and to what audience?
Also thanks for mentioning Ericsson's OF 1.1 work. Let me clarify what "hacked" means, because it might sound too frightening for some :o). Basically 1) there is no python support, and 2) the OF event handling and packet sending is replaced with a new one, which uses some C-structures. For examples, see [1,2]. I believe this is good enough (tm) for anyone wanting to prototype; and once a proper NOX comes out, migrating to that will be a non-issue. Be warned though, that due to the lack of a existing 1.1 test environments, both the 1.1 software switch and NOX might contain bugs. Nevertheless, it is our plan to migrate our existing MPLS demonstrations to these, so you can expect support from us - and from me personally. Any feedback is appreciated from those bold enough to give them a try =-). Regards, Zoltan. [1] https://github.com/TrafficLab/nox11oflib/blob/master/src/nox/coreapps/switch/switch.cc#L125 [2] https://github.com/TrafficLab/nox11oflib/blob/master/src/nox/coreapps/switch/switch.cc#L166 ________________________________________ From: [email protected] [[email protected]] On Behalf Of Rob Sherwood [[email protected]] Sent: Wednesday, August 24, 2011 5:19 AM To: [email protected] Subject: Re: [openflow-discuss] Slides from talk about OpenFlow 1.1 changes Opps .. attached the keynote version -- pdf included below. [Thanks to Srini for pointin this out] - Rob . _______________________________________________ openflow-discuss mailing list [email protected] https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
