Tuesday, July 24, 2007
New models for networking
"A new way to look at Networking" is a presentation by Van Jacobson about the next generation of networking. It is worth viewing on Google Video.
Van Jacobson could be described as the guy who saved the Internet from congesting, by inventing the slow start algorithm for TCP/IP. This presentation goes deep into the fundamentals of digital communication, yet presents a very clear vision of future networking. He gives no answers, but poses the right questions.
Jacobson states that the shift from the phone system to the Internet was a huge change, and was the result of a major paradigm change from circuit switching to packet switching. He then states that our current problems in networking can only be solved by changing our paradigm again, this time to 'dissemination'. The Internet is no longer mostly about conversations between end-points, but largely about dissemination of content.
This paradigm will allow better security, better mobility, and better use of broadcasting technology to avoid unnecessary inefficiencies, to name just a few. Early examples of this paradigm in action can be found in content distribution networks and peer to peer networks.
Van Jacobson could be described as the guy who saved the Internet from congesting, by inventing the slow start algorithm for TCP/IP. This presentation goes deep into the fundamentals of digital communication, yet presents a very clear vision of future networking. He gives no answers, but poses the right questions.
Jacobson states that the shift from the phone system to the Internet was a huge change, and was the result of a major paradigm change from circuit switching to packet switching. He then states that our current problems in networking can only be solved by changing our paradigm again, this time to 'dissemination'. The Internet is no longer mostly about conversations between end-points, but largely about dissemination of content.
This paradigm will allow better security, better mobility, and better use of broadcasting technology to avoid unnecessary inefficiencies, to name just a few. Early examples of this paradigm in action can be found in content distribution networks and peer to peer networks.
