Background

Scott joined ORNL in 2011. He is a Senior Research Scientist who leads the System Architecture team. His team of system architects and system programmers focuses on compute, memory, and interconnect technology trends to understand how to optimize existing systems such as Summit, how to prepare for next generation systems like Frontier, and will define the requirements for the system that will follow Frontier. His interests include resource heterogeneity within processors, within nodes, and across the system; processor architectures (e.g., CPU, GPU, FPGA, coarse-grained reconfigurable arrays); memory architectures and hierarchies; persistent memory; interconnects (e.g., PCIe, CXL, Gen-Z, InfiniBand, Ethernet); system scheduling; resilience; and system monitoring.

For Oak Ridge Leadership Computing Facility’s (OLCF) Frontier project, Scott is the Technical Project Officer (TPO). Scott led ORNL’s hardware evaluation of the CORAL-2 proposals. After ORNL and LLNL selected Cray, Scott was part of the team that negotiated the contract for Frontier.

Scott has been very involved in DOE’s Exascale Compute Project. He served as DOE’s Technical Representative for AMD’s FastForward-2 Node Architecture project and for AMD’s PathForward project. He also serves as the lead for ECP’s HPCM and Slingshot Test and Evaluation effort.

Education

2002
University of Tennessee
Computer Science
Master of Science (M.S.)
1987
University of Tennessee
Business Administration
Bachelor of Science (B.S.)

R&D Activities Contributions

Exploiting Node-Local, Non-Volatile Memory (NVM) - Spectral is a transparently applied library for taking advantage of the Summit Burst Buffer architecture. Applications using per-process output simply write to the node-local burst…

Reliability and Resiliency - The project involves analysis of the reliability characteristics of Titan’s 299,008 CPUs and 18,688 GPUs to understand trends in machine failure, MTBF, single bit errors,…

HPC Systems Scheduling Improvements - Resource selection can have profound impacts on the performance and reliability of applications running on the supercomputer. On Titan, there are on going efforts to…

DDSD: Data-Driven HPC System Design - As supercomputers grow more expensive, a more data-driven procurement strategy is needed to optimize the overall system performance while avoiding speculative resource provisioning at the…

Highlights