Google, SCE professor improve KC-area network
On March 30, the entire country watched as Google named its first high-speed broadband network site — Kansas City, Kan. At UMKC — just seven miles east of Google’s national test site — Deep Medhi also is working to make Internet communications better, faster and more personalized.
Because of his related research, Medhi — professor of Computer Science and Electrical Engineering and UMKC Trustees’ Faculty Fellow in the School of Computing and Engineering (SCE) — answered questions about the #Google Fiber project during an NBC Action News live chat. Click here to read the chat transcript.
The GENI project
If he had to summarize his research, Medhi would say he is specifically concerned with how the Internet will serve people’s needs in 20 or 30 years. Throughout the U.S., other researchers are wondering the same thing. Out of freewheeling conversations among computing and engineering scholars, the Global Environment for Network Innovations (GENI) was formed to explore the possibility of changing the Internet at its very core. GENI is a multi-site virtual laboratory for network science and engineering research, funded by the National Science Foundation.
As part of this effort, UMKC’s member group, the Great Plains Environment for Network Innovation (GpENI), received a three-year grant worth $462,500. Medhi is trying to devise a programmable network. GpENI partner schools each have their tasks: the University of Nebraska is working on optical networking needed for a programmable network. Kansas State University is working on the end device programmability, and the University of Kansas is devising a new control to connect these pieces.
The Sandbox
The Great Plains Environment for Network Innovation is a collaboration among UMKC, the University of Kansas, Kansas State University, the University of Nebraska–Lincoln, Ciena, Qwest, the Kansas Research and Education Network and Missouri Research and Education Network.
With their portion of the funding, Medhi and his student assistants operate a test bed that fulfills two roles. Thinking of the test bed as a sophisticated sandbox filled with familiar toys, researchers must bring these items together in different ways to make communications better, faster and more personalized. Next, they must test their findings – the experimental component. All this has to be done while maintaining the stability of the system.
Medhi’s lab in Flarsheim Hall houses the sandbox as well as his other computing equipment. He is quick to praise Mary Lou Hines, vice provost and chief information officer, as well as the Information Services staff for creating what he calls a “back road” – a wired route that runs parallel to his office computer setup. He can perform his normal computing tasks and enter the test bed seamlessly from the same lab. That way, Medhi’s lab is always available to him and testing stays clear of campus processes.
Above is a map of the Great Plains Environment for Network Innovation infrastructure.
Medhi carefully distinguishes between an “all new” system, which may not be neccessary, and a recalibrated existing system that does everything on the wish list. That raises the central question for Internet theorists – should everything be scrapped, the slate wiped clean and a new system born of revolution? Or is there a logical evolution from where the Internet is now to where it could be?
A Web-based demo welcomes others to tinker in the sand box. Interested researchers and individuals have accepted the offer, some from Europe and Asia, some as close as UM-Columbia and Missouri State University, intrigued by the excitement and promise of this line of inquiry. With people of various experiences and backgrounds diving right in, Medhi said there is no way to know where the next new thing will arise.