Sunday, December 27, 2009

Clean Slate Internet

The Clean Slate Program is an interdisciplinary research program at Stanford University which aims to consider how the internet would be redesigned with a "Clean Slate". Its Program Director is Nick McKeown. 
They identify five key areas for research:
  1. Network architecture
  2. Heterogeneous applications
  3. Heterogeneous physical layer technologies
  4. Security
  5. Economics & policy



Clean slate design: thoughts

Different communication paradigms
1. Phone networks were about wires, Internet about communication and networking of users, the Future
Internet is more and more about sharing of user generated content
2 The network itself is becoming more and more a large distributed database
3 The push and pull paradigm is changing due to the increase of storage in the network, which mediates the communication between users
4 Enable applications to move functionality inside the network: changing the end-to-end paradigm
5 Maybe multiple architectures are needed to consider different requirements at the same time:
            􀂁 Anonymity and accountability and security
            􀂁 Bulk data transfer and real-time communication
            􀂁 Performance and functionalities
6 Internet has no built-in security mechanisms, because it relies on cooperation and trust
􀂈 Internet is more and more about wireless access
􀂁 Spectrum allocated to Internet access is only a tiny fraction – most spectrum is unused
􀂁 Integration of wired/wireless networks
􀂁 Seamless mobility of users/services/networks

The design goals underlying the current Internet architecture in order of importance are:
(0) to connect existing networks,
(1) Survivability,
(2) To support multiple types of services,
(3) To accommodate a variety of physical networks,
(4) To allow distributed management,
(5) To be cost effective,
(6) To allow host attachment with a low level of effort and,
(7) To allow resource accountability.

Clean Slate Research Projects:

1.      The Programmable Open Mobile Internet (POMI) 2020 Project

2.      OpenFlow

3.      Web Security Projects
4.      Programmable Virtual Infrastructure for Virtual worlds
5.      Clean Slate Approach to Wireless Spectrum Usage

6.      A Clean Slate Approach to Enterprise Network Security: Ethane

7.      The NetFPGA Program
8.      Flow-level Models for the Internet
9.      Fast Dynamic Optical line paths for the Internet core 
10. Rate control Protocol

No comments:

Post a Comment