Trey White

Trey White

Distinguished Research Scientist

James B White III (Trey) is a performance engineer who strives to enable and accelerate science breakthroughs on advanced computers in the National Center for Computational Sciences. He has decades of experience in high-performance computing across a broad range of applications and technologies.

Oak Ridge National Laboratory
Distinguished Research Scientist, Jan 2024 - present

Hewlett Packard Enterprise
Master Engineer, Jan 2019 - Jan 2024

Senior Staff Software Engineer (HPC), Nov 2015 - Dec 2018

Children's Hospital Los Angeles
Applications Architect, Aug - Nov 2015

DreamWorks Animation
Senior Software Engineer, Apr 2012 - Aug 2015

National Center for Atmospheric Research
Senior Software Engineer, Jun 2010 - Apr 2012

Oak Ridge National Laboratory
Senior Research Computer Scientist, Mar 1999 - Jun 2010

Ohio Supercomputer Center
Computational Scientist, 1997 - 1999

University of Tennessee, Knoxville
PhD in Computer Science, 2011
Dissertation: "Algorithms for Advection on Hybrid Parallel Computers"
Advisor: Jack Dongarra

The Ohio State University
MS in Physics, 1996

Rhodes College
BS in Physics with Honors, 1992

Performance, Portability, & Productivity in HPC
Program Committee, 2020-present

University of the People
Online Adjunct Instructor, 2017-2019

League of Amazing Programmers
Volunteer Teacher, 2016-2017
Robot Competition Team Mentor, 2016

SC Conference Series
Tutorials Committee, 2001-2002, 2006, 2011-2014
Chair of HPC Software Challenge, 2004
Technical-Papers Committee, 2003
AV Chair, 2000

International Journal of HPC Applications
Subject-Area Editor, 2010-2012

Oak Ridge Leadership Computing Facility
User Council, 2010-2012

Swiss National Supercomputing Centre
Scientific Reviewer, 2009-2012

Cray User Group
Local Arrangements Chair, 2004, 2009
Board of Directors, 2004-2007

IBM HPC Systems Scientific Computing User Group
Program Committee, 2006
Founding Member, 1999

then the Compaq Alpha Scientific and Technical User Group
Founding Member, 2000

2016-2018, 2023-present

Performance Versus Maintainability: A Case Study of Scream on Frontier

Performance Portability of Programming Strategies for Nearest-Neighbor Communication with GPU-Aware MPI
The Simple Cloud-Resolving E3SM Atmosphere Model Running on the Frontier Exascale System
Experiences Readying Applications for Exascale
Exploring Fully Offloaded GPU Stream-Aware Message Passing
Designing HPE Cray Message Passing Toolkit Software Stack for HPE Cray EX supercomputers (paper) (slides)
Application Profiling (slides) (video)

Exploring GPU Stream-Aware Message Passing using Triggered Operations
Debugging and Performance Profiling for Frontier (video

A Comparison of GPU Programming Models (slides) (video)

Relative Outcomes of Climate Change Mitigation Related to Global Temperature Versus Sea-Level Rise
Climate System Response to External Forcings and Climate Change Projections in CCSM4
A Modern Solver Interface to Manage Solution Algorithms in the Community Earth System Model

High-Performance High-Resolution Semi-Lagrangian Tracer Transport on a Sphere
Overlapping Computation and Communication for Advection on Hybrid Parallel Computers
Multiwavelet Discontinuous Galerkin-Accelerated Exact Linear Part (ELP) Method for the Shallow-Water Equations on the Cubed Sphere

A Scalable and Adaptable Solution Framework within Components of the Community Climate System Model
Time Acceleration Methods for Advection on the Cubed Sphere

Acceleration of Time Integration

Software Design for Petascale Climate Science
Scientific Application Requirements for Leadership Computing at the Exascale
It’s About Time: Multi-Resolution Timers for Scalable Performance Debugging

Comparing Optimizations of GTC for the Cray X1E and XT3

Systematic Study of d-Wave Superconductivity in the 2D Repulsive Hubbard Model
Vectorizing the Community Land Model
Practical Performance Portability in the Parallel Ocean Program (POP)
Performance Evaluation of the Cray X1 Distributed Shared-Memory Architecture

Adventures in Vectorizing the Community Land Model
Dangerously Clever X1 Application Tricks
Towards Full Simulations of High-Temperature Superconductors
ORNL Cray X1 Evaluation Status Report

Early Evaluation of the Cray X1
An Optimization Experiment with the Community Land Model on the Cray X1
DOE Ultrascale Evaluation Plan of the Cray X1

Early Evaluation of the IBM p690

Where's the Overlap? An Analysis of Popular MPI Implementations

On Improving the Performance of Sparse Matrix-Vector Multiplication
Performance Issues of Scientific Programming in Ada 95