I’m intrigued and a bit perplexed by the recent announcement of the Open Networking Foundation and its vision of programmable networks. This quote crystallizes what I’m wondering about:
“In the OpenFlow network, all the intelligence will be in a central point, so it’s easier to do complex things.” — Urs Hoelzle, SVP of Engineering, Google
I thought the Internet was all about de-centralized intelligence?
I know ONF is a serious effort built on years of research work and with support from major industry players. I’m not religious in believing everything must be decentralized and end-to-end. My latest paper explains how cloud computing represents a shift from the Internet model back to an earlier vision of centralized platforms. That being said, either this effort is a way to bridge some of the limitations of the Internet model and traditional centralized networks… or it’s a challenge to the Internet model. Either path is interesting.
Ironically, they’re using a name we developed in a completely different way (independently) at Supernova 2008. The networking protocol dates back to the same period, so I’m not objecting. I would merely question whether the networking project is fundamentally about openness and flow. It’s open in that it’s an industry consortium, but so far not through an independent standards body. I wonder whether the small, innovative players will like the approach as much as the big data center operators driving the effort.
And how does centralized intelligence promote flow? To me, flow is a state, not a state manager. When we conceptualized Open Flow in business at Supernova, we were talking about getting out of the way of the data stream and harvesting it to create business value, not figuring out how to manage and control it efficiently.
All this points to the likelihood that the Internet as we’ve known it is coming to an end. That’s not necessarily a bad thing, although it could be. The danger is that those of us on the right side of history in the Nethead/Bellhead debate will be on the other side of the Nethead/Nexthead debate. Worth watching closely how things unfold.