What is advantage of using proactive open flow controller over using remote command(like a, running ovs-ofctl with ssh)? I read http://docs.projectfloodlight.org/display/floodlightcontroller/Static+Flow+Pusher+API. And It look like that i need to know switch id and port id if i want to insert openflow flows. But how can i proactively know these id only with openflow controller? Maybe i can insert open flow flows with this API. But if i don't know where the switch is (in topology sense) and what is id of the node and the port, how can i know which rule applied to? I read Open Stack network(Qunatum) source code. It proactively inserts flow with remotely calling special daemon agent which gather local information(ex ID) and run local command(like ovs-ofctl). How can proactive openflow controller get advantage over these remote daemon?
_______________________________________________ openflow-discuss mailing list openflow-discuss@lists.stanford.edu https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
