Consortium outlines goals for advanced, user-friendly Internet2 infrastructure
- By William Jackson
- Jul 11, 2008
The Internet2 consortium has adopted a new five-year strategic plan focused on keeping the national research and education network at the cutting edge of performance while making it easier for nonexperts to use.
'It's a natural leap, focusing not on the bits but on how the bits are used,' said Peter Siegel, chief information officer at the University of California, Davis, and chairman of the Internet2 Steering Committee's Research Advisory Council.
The Internet2 community also wants to encourage a national telecommunications policy to support an infrastructure for new communications needs, such as telemedicine.
Getting federal support 'is probably going to be a challenge,' Siegel said, adding that the community needs to define the business case for such a policy.
Internet2, established in 1996 to pick up where the NSFnet Internet backbone left off, is a test bed network for new technologies and applications. 'We provide a breakable test bed' on which users can try new approaches that might not be feasible on commercial networks, said spokeswoman Lauren Rotman. Internet2 also provides bigger pipes to accommodate the needs of power users in the research community. A consortium of 212 universities, 70 corporations and 45 affiliate members operate the network.
The consortium's activities have expanded to include advanced network applications, middleware, tools and services.
The Internet2 Board approved the strategic plan July 9. The result of an eight-month effort by an executive committee, it outlines a core mission to provide scholars and researchers with the advanced networks, tools and support required for the next generation of collaborative work and innovation. Accomplishing that goal will require moving beyond the high-speed services now available on commercial networks.
'There is an opportunity to tie the computing infrastructure more closely to the application,' Siegel said. 'We have not yet built from the network into the application layer.'
Once that happens, it will allow advanced applications to use the network more effectively and tap information already in the network to help recognize and accommodate applications' characteristics and needs. It will also require more intelligence on the part of the network and new middleware to mediate for the applications.
'There is a very clear need for strengthening the role of middleware -- the glue between the application and the infrastructure,' Siegel said.
The basic functionality exists, but many typical users ' both researchers and students ' cannot easily take full advantage of it to set up ad hoc collaborative environments using advanced applications at different locations.
'That part needs to become automated to make it accessible to people who are not networking experts,' Siegel said.
Internet2 also collaborates with advanced research and education networks in other countries, many of which are supported by national telecommunications policies.
'In the U.S., there isn't really that kind of role' for the government, Siegel said. 'To some extent, it has been a disadvantage,' forcing the research and education community to do the work itself. But it has also produced a broad, active community with the freedom to innovate.
The challenge will be to create a policy with government support for research and deployment without squelching the community's creativity.
'We have to take a leadership role in defining what the program needs to be,' Siegel said.