UMass Amherst Researchers Plan to Bring More Freedom of Choice to the Future Internet

When the Internet was designed nearly 50 years ago, it was hard to imagine that one day nearly everyone would be browsing, banking, shopping and streaming entertainment on it. Due to such broad and diverse use, today it faces security, bandwidth and quality of service problems, say computer engineering researcher Tilman Wolf and operations management expert Anna Nagurney at the University of Massachusetts Amherst.

They recently received a three-year, $909,794 National Science Foundation grant to address some of these difficulties with new protocols and services. The project, "Network Innovation through Choice," is part of a $2.7 million collaborative that includes UMass Amherst, the University of Kentucky, North Carolina State University and the Renaissance Computing Institute of Asheville, N.C.

As Wolf, an associate professor of electrical and computer engineering, explains, NSF has supported many such projects on network architecture and prototyping to foster a national discussion plus practical experiments among top computer scientists and other researchers about what a new Internet might look like and how it would be built. The UMass Amherst project will use an NSF-funded Global Environment for Network Innovation (GENI) testbed already used by other researchers on campus.

"In the near future, there will be many different ways of communicating on the Internet," says Wolf. "One thing missing is that current technology doesn’t give users any control over the many choices to come. For example, you typically must choose a single Internet service provider (ISP) and you are stuck with the network performance they provide."

"What if you could choose an ISP in your area with a button on your desktop and you could pay for only the time you used each one? To watch a movie with the fastest service for high-quality streaming, you buy it for two hours and move on. For checking your e-mail later, you take the lower-priced product that offers higher security, a secure server with the latest anti-virus software. That’s the sort of choice and flexibility we’re talking about."

He and Nagurney, director of the Virtual Center for Supernetworks at the Isenberg School of Management at UMass Amherst, recently were chosen to lead the multi-institution project intended to explore the design and development of next-generation network architecture, in essence to envision a new type of Internet. They are particularly interested in offering users many more choices than are available in the present cyber environment and in letting market forces reward the most desirable.

Nagurney, who is also the John F. Smith Memorial Professor of operations management, explains, "We’ll create a framework that will reward innovation based on the economics of choice. With our network design expertise, we can develop systems that allow consumers, developers and other users to reward more choice on the Internet through payment and reputation. For example, broadband performance and improvement will become more transparent and accountable to consumers as they choose services. Our goal is to bring similar incentives to as many Internet innovations and services as possible. We want to offer a level of customization for individual needs."

Wolf and collaborators will tackle the technical side of the two-part problem while Nagurney will take on the economic and business side. In three years, they hope to be able to demonstrate a working, open source prototype of a new network using a testbed where users can have access to and choose from many different types of services and functions on the network.

"The underlying technologies will run invisibly to the user," Wolf says. "So the computer won’t ask if you want 200 millisecond round-trip time and no packet loss or 180 millisecond round-trip time and 1 percent packet loss. Rather, the software would just ask if you want to pay three cents an hour more for a better movie experience."

Overall, he says, "The view now is that in the future there will be many different networks or Internets in parallel available to users who want different network characteristics. People will switch between them for different uses. So the banking system might switch to a much more secure network and you won’t be able to access your account unless you join that."

"There are many, many questions to be resolved such as who owns, who regulates, and who pays for these new networks," he adds. "Many of the answers will likely come from such projects as the one we’re starting."

Provided by University of Massachusetts Amherst