Barker studied computer science at North Carolina State University and the University of Notre Dame before earning his PhD in 2004 from the College of William and Mary in Virginia in the area of runtime software tools to support adaptive and dynamic large-scale parallel scientific applications. He began his career at Los Alamos National Laboratory, where he worked on developing models to describe large-scale scientific application performance on current and future supercomputing architectures. These models were used to drive procurements and guide workload optimizations to enable scientific codes to achieve greater performance, scalability, and efficiency on high-performance computing systems.
Moving to PNNL in 2010, Barker currently leads the High-Performance Computing group. He currently researches advanced computing architectures and hardware/software co-design methods and leads the CENATE project, which explores the impacts of emerging and novel technologies on the Department of Energy (DOE) computing mission and is funded by DOE’s Advanced Science Computing Research program. Barker also serves as the thrust leader for Heterogeneous Computing within the Data-Model Convergence (DMC) Initiative.
Disciplines and Skills
- High-Performance Computing
- Computer Architecture
- Performance Modeling and Analysis
- Runtime System Software
- Programming Models
PhD in Computer Science, College of William and Mary, 2004
MS in Computer Science, University of Notre Dame, 2001
BS in Computer Science, North Carolina State University, 1997
Affiliations and Professional Service
- Association of Computing Machinery