January 14, 2021
By Molly Sharlach, Office of Engineering Communications
Princeton University researchers are partners on a major initiative to help shape the next generation of computer networks. The effort, known as Project Pronto, aims to build and test new types of flexible, programmable networks to advance network security, performance and innovation.
As fifth-generation (5G) wireless networks connect a growing number of computers, phones, sensors and other devices, the ability to customize network infrastructure could improve both security against cyberattacks and the safety and performance of devices such as drones and autonomous vehicles.
Project Pronto is supported by a three-year, $30-million grant from the U.S. Defense Advanced Research Projects Agency. Leading the Princeton arm of the project is Jennifer Rexford, chair of the computer science department and an expert on network routing, measurement and management. The project also includes collaborators at Stanford University, Cornell University, and the nonprofit Open Networking Foundation.
The project is one of the largest U.S. government investments in networking since the creation of ARPANET, which launched in 1969 and was the precursor to the modern internet. “The scale of the project is noteworthy,” said Rexford, Princeton’s Gordon Y.S. Wu Professor in Engineering. “We want to move the needle on how the internet works.”
The critical need for timely communication with systems that control drones and self-driving cars will put new pressures on wireless networks, said Rexford. Growth of these kinds of cyberphysical systems “forces us to grapple with every challenge amped up: more scale, more mobility, more need to react quickly,” she said.
Currently, many aspects of computer networks are predetermined by their equipment and software, which network administrators have limited ability to adapt. Project Pronto aims to create new tools that allow administrators to tailor networks to their needs, as well as more nimbly detect and mitigate problems or potential attacks.
“We want the owners of the network to have the ability to change how the network functions,” Rexford said. In building new 5G networks, “we think it’s important that we get it right before the door closes.”
The three university campuses will serve as testbeds for the project’s programmable networks. At Princeton, researchers will experiment with network hardware installed in the Friend Center.
“Project Pronto envisions an innovative ‘campus as lab’ infrastructure that will demonstrate how to provide secure wireless communications between teams across multiple universities,” said Jay Dominick, Princeton’s vice president for information technology and Chief Information Officer. “This research project will inform how the Office of Information Technology can creatively deploy a high-speed, campus-wide 5G network to meet Princeton University’s future wireless communications needs.”
The project is “a way for us to dip our toes into what the future networking infrastructure might look like,” said Rexford, adding that programmable networks will be key to “the campus of the future,” which will likely include smart sensors and autonomous vehicles.
Her group plans to collaborate with other Princeton researchers who focus on robotics and different aspects of wireless networking. “As we get this up and running, my hope is that we start to build a community of people who will be users of this infrastructure,” she said.
Rexford is among the creators of a network programming language called P4, which will be used by Project Pronto researchers to program network devices and measure network activity. The programmable networks will also operate using the open-source platform Aether, developed by the Open Networking Foundation.
The grant’s principal investigator is Nick McKeown of Stanford. Other partners include Nate Foster of Cornell; and the Open Networking Foundation’s Guru Parulkar, Oguz Sunay and Larry Peterson. Peterson is the foundation’s chief technology officer and Princeton’s Robert E. Kahn Professor of Computer Science, Emeritus.