Skip to content Skip to navigation

Linking Research and Patient Care: the University of Florida’s Bold Internet2 Adventure

July 5, 2013
by Mark Hagland
| Reprints
Gigi Lipori and Erik Deumens, Ph.D. describe their work linking the University of Florida and the University of Florida Health through advanced connectivity for improved patient care

A particularly exciting development in the world of information technology has been the creation and forward evolution of Internet2, a community of U.S. and international leaders in research, academia, industry, and government, who are collaborating to create new channels for communication in order to support research and development.

As its website indicates, the collaborative known as Internet2 encompasses 220 U.S. universities, 60 leading corporations, 70 government agencies, 38 regional and state education networks, and “more than 100 research and education networking partners representing over 50 countries.”

Among the U.S. universities that is farthest along with regard to connecting genomic research and patient care delivery is the University of Florida at Gainesville, where developers have been connecting genetic and genomic research with cancer treatment. As part of the technology backbone for this work, the University of Florida folks and other Internet2 collaborative partners are being supported by the San Jose, Calif.-based Brocade.

Recently, HCI Editor-in-Chief Mark Hagland spoke with Gigi Lipori, MT, MBA, the University of Florida Health’s senior director for operational planning and analysis,  and Erik Deumens, Ph.D., the University of Florida’s director of research computing, about the work that they and their colleagues are currently involved in. Below are excerpts from that interview.

Can you share with me a bit about the origins of Internet2?

Erik Deumens, Ph.D.: When the Internet was invented decades ago, it grew out of an academic and military background. But in the beginning of the 2000s decade, universities got together and said, we’ve got this Internet, but we can no longer do research on it anymore. So they created a new consortium with 200 members, and called it Internet2, for second-generation Internet. And the first focus of activity was to try to get faster speed.

So of the more than 400 organizations involved, how many are universities?

About 150 of them; the rest are companies, like Intel and Dell, as well as some research institutes like the Broad Institute, a group of people with interest in moving data. The Internet2 collaborative is focused on increasing speed; what has happened in the last two years is that Internet2 received a grant from the federal government to create a 100-gigabit-per-second backbone. The typical Internet connections are 10 or 20 gigabits between cities. So our collaborative came up with the concept of the Internet2 platform. Currently, about 26 universities and research institutions have committed to become an Internet2 innovation platform.

There are three conditions: first, you have to get from wherever you are to the closest backbone site. In the case of the University of Florida, we have to make a connection between Gainesville, where the university is, and Jacksonville, the nearest site. The second condition is to have a “science DMZ.” A DMZ (demilitarized zone) is a separate network—it means that the typical security people allow certain things to happen—the researchers, if they have a particular project, there’s a quick, separate approval process. Because typically, an institution is protected by strong firewalls; so the firewalls are lowered somewhat, so the research can keep on happening. And the reason for doing this is that if you tried to send a terabyte of data from one university to another, and it had to go through a firewall, and the firewall had to inspect every packet, that would be way too slow.

So if you know from the beginning you were going to send a terabyte data set, all belonging to the same stream, and if you inspect the first packet, you can say, the whole data set is OK. And the third requirement is that you have to do active research in software-defined networks—a new technology where you’re trying to create more efficient protocols. Currently, things are governed by the IP—Internet protocol. And if you send a packet, the first couple of bytes have the format of the address in them, and every router across the Internet has to read this address information; but we all know that that protocol was invented 30 years ago, and we really would like to make that more efficient. The Internet is very stable; and has never crashed as a whole yet. But we all know that the Internet protocol is dated, and we’d like to test a new protocol; and you certainly cannot test it on the current production network, because it would create chaos. So with software-defined networks and DMZs, you’re trying to create a separate platform where researchers are trying to innovate.

So then what happened is that the University of Florida made an investment and got a grant. And we were trying to be early adopters; it turns out the University of Florida is the first university in the nation and the world to meet all three criteria of the innovation platform. And Internet2 was pretty excited, and they issued a press release, and said, we want to explore other areas the university is working on, in terms of the use of its science DMZ for healthcare. So the University of Florida—and I made this statement at one of the meetings—the University of Florida will now build a storage infrastructure that is HIPAA-compliant, that will allow people to do research and use fact connectivity for genomic and protected health information.

So how far are you on the healthcare side?