The Defense Advanced Research Projects Agency, the research and development arm of the U.S. Department of Defense, has awarded a three-year, $30 million grant to a team including Cornell researchers to build a fully programmable computer network, in order to bolster internet security and help fuel innovation.
Networks allow computers to communicate with each other, but consolidation in the technology industry has led to error-prone and inflexible networks that are difficult to customize. Programmable networks can allow operators greater control, because they don’t have to reply on vendors and can build and deploy more secure and reliable systems.
Though recent advances in technology could allow operators to program their own networks, little has been done until now to address the security and robustness of computer infrastructure.
“We are quickly moving to an era where network behavior is defined by software engineers in code. This change brings tremendous potential, but also a need to ensure the network is behaving as intended,” said , associate professor of computer science and a co-investigator on the grant.
“Our goal is to create automated verification tools that can help identify unintended anomalies and secure networks against intentional attacks,” Foster said.
The grant’s principal investigator is Nick McKeown, professor of electrical engineering and computer science at Stanford University. Also partnering on the grant are Jennifer Rexford, professor of computer science at Princeton University, and Guru Parulkar, executive director of the Open Networking Foundation and executive director of the Stanford Platform Lab.
The project, which was awarded the grant in the spring, launched in December.
The researchers from Cornell, Princeton and Stanford are working on a project known as Pronto, which will build a platform supporting fine-grained measurement, networking and closed-loop control. Their goal is to ensure the integrity of the network and reduce the possibility that human error or bad actors can disrupt it.
