Name: Damian Rouson
Pronouns: he/him/his
Biography:
Damian Rouson is a Staff Scientist and the Group Lead for the Computer Languages and Systems Software (CLaSS) Group at Berkeley Lab. He is a mechanical engineer with experience in simulating turbulent flows in multiphase, quantum, and magnetohydrodynamic media. He leads the development of the OpenCoarrays and Caffeine parallel runtime libraries and the Inference-Engine deep learning library. His work at Berkeley Lab involves researching ways to accelerate predictions of climate change’s regional impacts, teaching UPC++, and contributing tests to the LLVM flang Fortran compiler. He co-authored the textbook Scientific Software Design: The Object-Oriented Way (Cambridge University Press, 2011) and has taught related university courses and tutorials on Fortran 2018 and agile software development. He has held staff and faculty positions at the City University of New York, the University of Maryland, the University of Cyprus, the University of Bergen, and Stanford University and staff and management positions at the U.S. Naval Research Laboratory and Sandia National Laboratories. He received a 2003-’04 NASA Summer Faculty Fellowship and a 2020-’21 DOE Better Scientific Software Fellowship. He founded Archaeologic Inc. and Sourcery Institute. He holds a B.S. from Howard University and a M.S. and Ph.D. from Stanford University, all in mechanical engineering.
Institution/Lab: Lawrence Berkeley National Laboratory
Website: http://go.lbl.gov/damian-rouson
SRP Collaboration Topic/Title: Language-based Parallel and GPU Programming for Deep Learning
Field or research area: AI for High-Performance Computing
Please select all the topical areas that apply to your project:
High-Performance Computing; Machine Learning and AI
Brief Abstract:
The goal of this project is to use language-based parallel and GPU programming to accelerate neural-network training in Berkeley Lab’s Inference-Engine deep learning library.
Desired relevant skills, background, or interests:
Programming in modern Fortran. Deep learning. Parallel or GPU programming.
Other comments:
Do any special requirements apply? U.S. Citizen Only; Permanent Resident OK; International OK; other
Other, specify: in-person or hybrid work preferred
Keywords:
Deep learning; Fortran; Parallel programming; GPU programming
Lightning Talk Title: Language-based Parallel and GPU Programming for High-Performance Computing