Paul Ruth
University of North Carolina at Chapel Hill, NC·Computer Science
Modern science moves huge amounts of data over networks, and the computer's main processor often becomes the bottleneck. SmartNICs are programmable network cards with their own processors.
- Fields
Computer Science
Engineering
Education Research
- Topics
SmartNICs
Data Processing Units
network interface cards
- Grants
- 2 known grants, through Sep 2027
What they work on
- Deploy next-generation SmartNICs, programmable network cards called Data Processing Units, at four international FABRIC testbed sites so researchers can offload networking, storage and security tasks from CPUs and GPUs.
- Enable data-intensive international experiments in high-energy physics, federated learning and wireless networking, with isolated workloads for secure networking and encrypted storage research on the FABRIC network.
- Develop web-based virtual labs, guided experiments and interactive booklets on FABRIC and a university cloud that teach developers how to program offloaded applications on SmartNICs using open-source tools.
- Train system administrators and support staff to deploy and manage SmartNICs, and build the virtual labs into community college, bachelor's and graduate curricula through workshops and train-the-trainer tutorials.
More about the research
The most advanced kind are called Data Processing Units (DPUs). They can take over tasks such as encryption, data filtering, storage management and even machine-learning inference, sometimes running orders of magnitude faster than the main processor. Both projects center on SmartNICs and on FABRIC, a national research network testbed. One adds SmartNICs to FABRIC's international sites for experiments. The other trains people to use them.
Why it matters
Scientists in fields such as high-energy physics and AI who move very large datasets could get results faster and more securely by offloading data handling to SmartNICs. Campus networks and small and medium-sized organizations, which have lagged behind large cloud providers, could adopt the technology more easily with accessible training. That could help with massive data transfers, fast machine learning and large-scale traffic inspection for cybersecurity.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Arvind Krishnamurthy
University of Washington, WA·Computer Science
Design NetFusion, a rack-level pool of programmable network interface cards (SmartNICs) that perform packet and network-function processing for many servers, with safe and fair sharing among applications.
Nikolai Sultana
Illinois Institute of Technology, IL·Computer Science
Develop remote data-plane attestation for P4-programmable network devices, generating verifiable evidence of their running program and state, with tunable trade-offs between integrity assurance and computing overhead.
Kenneth Calvert
University of Kentucky, KY·Computer Science
Determine where and how to connect Named-data Networking (NDN), which fetches data by name, inside a network domain with SCION, a secure routing system that carries traffic between domains.
The school
University of North Carolina at Chapel Hill: 72 labs on PI Pages, $1.6B a year on research, 523 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.