Ivan Pepelnjak has written a great reaction to what might become a thread started by Brad Hedlund and Joe Onisick pulling everybody back to the ground again on network virtualization. Something he does more often.
Some of those products will be well executed and highly profitable, but in the end all of them will go down the same path as X.25, ATM and LANE went a decade ago, because they’re architecturally suboptimal.
I think Ivan is right with this conclusion about the numorous overlay network solutions currently popping up, but I am afraid this also counts for SDN and especially Openflow. Although Openflow does not have to keep track of state, it certainly has to manage every flow.