Internet may be showing its age

Hitting the big 4-0 might prompt thoughts of a makeover, and that’s true for the Internet as well, according to a Stanford University professor who earned his doctorate in computer science at UD.

Guru Parulkar, AS ’87PhD, a consulting professor and executive director of the Clean Slate Internet Design Research Program at Stanford, recently told an audience on the UD campus that now is the right time to reinvent the Internet.

Parulkar described a new initiative for that redesign in his keynote speech, “A Case for Rethinking the Internet Architecture,” to an audience of more than 150 students, faculty, staff and computing science professionals. The Feb. 22 talk was part of UD’s inaugural Computer Science Research Day.

“The Internet has been wildly successful and has been an important invention for the benefit of society, but its architecture is nearly 40 years old,” Parulkar said. “In the years since its invention, new uses and abuses have pushed it into realms that its original design neither anticipated nor was able to easily accommodate.”

In designing a new Internet, he noted, researchers must consider such issues as control management, addressing and naming the inter-domain routing, mobility of hosts and networks and economic viability of stakeholders. He said another limitation of current Internet architecture is that it was not designed to absorb innovations, including scaling, or increasing, of services.

“Google built one of the largest infrastructures in the world,” Parulkar told the audience. “Today, everyone has to build their own infrastructure. The problem is that to turn your idea into service, you have to integrate it into the current capability of the Internet architecture.”

He described the Clean Slate initiatives, which include creating a new Internet that would solve the current deficiencies and allow it to become a unified global communication infrastructure. The proposed Clean Slate program will focus on unconventional, bold and long-term research that tries to break the network’s ossification and overcome architectural limitations, including lack of security and mobility, he said.

“We need to ask ourselves, with what we know today, if we were to start again with a clean slate, how would we design a global communications infrastructure,” Parulkar said. “We also have to ask ourselves how the Internet should look in 15 years. We will measure our success in the long term. We intend to look back in 15 years’ time and see a significant impact as a result of our program.”

He acknowledged that programs like Clean Slate and the Global Environment for Network Innovation (GENI), a national-scale facility for conducting networking and computer science research, have their opponents as well as supporters.

“These ideas are not universally admired. Some say they are too broad, and other say they are not broad enough,” Parulkar said. “Maybe these views are a reflection of how the computer science community still does not know how to champion such multi-funding initiatives.”

In recalling his days as a graduate student, he urged potential doctoral candidates to choose UD as the place to pursue their advanced academic goals while becoming part of the exciting potential of future research efforts.

“I got a great education at UD, and I would encourage you to choose UD for graduate school,” Parulkar said. “Our job is to enable you to build new structures. This is really an exciting time to be graduate students.”

Before becoming director of Clean Slate, Parulkar served as an assistant professor at Washington University in St. Louis, where he and two colleagues designed and implemented a new type of communication system that permits Internet routers to step up and handle optical network speeds. The company, Growth Networks, was purchased by Cisco Systems in April 2000 for $335 million.