PNNL

High-performance computers are now critical to scientific endeavors across diverse disciplines such as biology, chemistry, climatology, and materials. At PNNL, computer scientist Kevin Barker researches the advanced architectures and hardware/software designs that enable these powerful machines to run both effectively and securely.

Barker co-leads the Center for Advanced Technology Evaluation (CENATE), a computing proving ground for emerging and novel computing technologies. For one project at CENATE, Barker and colleagues developed a system to detect questionable use of high-performance computing systems within the Department of Energy. The system is one part of a larger mission to keep the nation's high-performance computers cybersecure.

“Machine learning methods are helping us to identify some of the key characteristics of workloads that represent legitimate scientific computing or something that might be anomalous that we would want to look into further,” Barker said of the cybersecurity project at CENATE.

Barker also leads a team that is part of the Data-Model Convergence initiative, a multidisciplinary effort to create the next generation of scientific computing. The Initiative's goal is to push computing farther by combining scientific modeling and simulation with machine learning and graph analytics.

“From an early age, I was drawn to computing and its potential for shaping the future,” he wrote on LinkedIn. “Enabling solutions to our biggest challenges in fields as diverse as climate, medicine, manufacturing, and agriculture; bringing people closer together through instantaneous global communication; and integrating into daily life to monitor and improve health care, enable autonomous vehicles, and increase energy efficiency fascinated me – this was a field I wanted to be a part of and contribute to.”

More Information

PNNL Staff Biography

LinkedIn Profile

“PNNL Researchers Unveil Tool to Accelerate CGRA Development.” June 18, 2021, HPCwire