Internet2: Bypassing the Internet

By Ayon Nandi

In 1969, helped by funding from the military, four computers were networked together in an effort to create a research network that could survive a military strike. That was the ARPANet. Soon after came the DARPANet. Now we have the Internet, perhaps the world's largest cooperative.

But it is getting crowded. At the same time, science today demands the ability to transmit much more data faster than before. "Everyone in the [science] world wants more bandwidth," Yale physics professor Michael Zeller said.

Scientists at Yale and dozens of other research centers will soon get more bandwidth through Internet2 (I2), a research-only network, separate from the Internet. In other words, it will be a second Internet: DARPANet all over again, only this time, it will be much faster, less crowded, and more efficient. This new network, the result of a collobarative effort among more than a hundred American research centers, will exclusively be used to shuttle data between scientis ts around the country.

"We ceased being able to use the Internet for science"

The researchers that began working with DARPANet in the early eighties find that the present Internet is too crowded with commercial sites, private users, and AOL users looking for cyber-dates for them to run the high-tech applications they need. Computer science associate professor David Nicol of Dartmouth University is one such researcher, and agreed that the present Internet's state is one of information bottlenecking.

"Before public awareness of the internet exploded, the traffic on it was low enough to be able to do things like use big expensive computers in a government lab, across the country. [Recently] we ceased being able to use the network for science. The traf fic on it became too great to provide necessary bandwidths," he said.

High-end researchers hope that I2 will be the solution to this problem.

Three goals

The goals of the I2 project, as spelled out on the project's web site, are threefold. First, the project, supported by the University Corporation for Advanced Internet Development (UC AID), a group of 120 US universities, hopes to "[create] and sustain a leading edge network capability for the national research community." This means directing efforts towards providing member universities and research groups with the hardware and soft ware necessary to support a new, high-bandwidth network.

I2 has already arrived at many universities, including Yale. Almost. The National Science Foundation (NSF), in July, awarded Yale a grant to help support its development of I2. Yale was one of 63 universities to get such a grant, which will connect them with NSF's own High Performance Backbone Network Service (vBNS). The connection, at 45 megabits per second (Mbps), will be 10 time faster than Yale's present 4.5 Mbps line, Zeller said. However, "the physical line has not yet arrived," he added.

The I2 group and UCAID have made the creation of applications for the new network the second part of their threefold plan. They need not worry; Yale scientists, who use high-resolution imaging, video conference with other scientists, and track a quark's path nanosecond by nanosecond have many ideas for how to use a high-bandwidth network. Zeller and three other professors at Yale won grants for proposals they submitted which would use I2, but, in practice, "once it's in here, we can distribute it, and ev eryone who needs it can use it," Zeller said.

Some professors already have specific plans for I2. Zeller looks forward to connecting from his lab on the fifth floor of the Gibbs building to the particle accelerator at Brookhaven National Laboratories, in Upton, New Yor k. Theoretical physicists at Yale will also benefit from I2 through an application the I2 group calls Tele-immersion, which would allow users to experience the effect of sitting in the s ame room with other scientists who are miles away. Diagnostic radiology professor C.C. Jaffe will use the network to transmit high-resolution medical images.

However, not all Yale researchers will be able to use I2, partly because not all of the world's science centers have access to the new network. Professor Charlie Baltay, chairman of the physics department, would like to use Internet2 for the QUEST project, which will use a telescope in Venezuela to look for stellar bodies known as quasars. However, only American universities are currently connected to I2. In an informal email survey of Y ale physics professors, many responded that they were not using I2, but had some interest in the new network. Applied physics chairman and professor A. Douglas Stone wrote in a recent email to Yale Herald Online, "I can imagine it could be important for my research at some point, but I haven't figured out how yet." Some other professors wrote that they had no plans to use I2 in the near future.

Will Internet2 give way to an Internet3?

Participation in the I2 project has thus far been limited to the US universities that have the capital and the resources to continually help in the development of new network applications and tools, as well as $25,000 a year for UCAID membership fees. Acc ording to University of North Carolina-Chapel Hill Systems & Communications Director Joel Dunn, "the participation to date is pretty much on target; it's ended up being the major computer and networking corporations and the first-tier research universitie s."

The Internet2, then, is very similar to the Internet of the late eighties and nineties. Back then, with very little Internet traffic, university groups developed a host of new, exciting applications that eventually became the basis for such things as the World Wide Web. Is such a future in store for Internet2? Will it, too, become overcrowded in ten years and give way to an Internet3?

Dartmouth professor David Nicol is concerned about that possibility. He said, "I am really concerned that without 'flow control', the state of Internet2 in three years will be no different than the current state of Internet1."

David Ward, chancellor of the University of Wisconsin-Madison and chair of the UCAID Trustees, emphasized that I2 remains an enterprise dedicated to research and education in centers of higher learning. "Our goal is to ensure that the educational and rese arch capacities [of I2] have the highest priority over pure information utility functions," he said.

While the I2 group plans to keep non-research groups out of the network, it will simultaneously work to get more scientists connected. In the third point of its threefold mission for I2, the web site states, "[we wish] to integrate the work of Internet2 w ith ongoing efforts to improve the production [of] Internet services for all members of the academic community."

For now, though, Internet2 remains in the laboratories of a few select researchers, providing new and exciting aplications for the twenty-first century.

What do you think? Respond in Speak your Mind.

Back to Online Features...