Mike Pozulp works at Lawrence Livermore National Laboratory.
He is also a PhD student at the University of California, Berkeley.
He may be contacted using the info in his resume.
Presentations and Publications
Lead author indicated by *
"Heterogeneity, Hyperparameters, and GPUs: Towards Useful Transport Calculations Using Neural Networks"
(with P. Brantley, T. Palmer, J. Vujic). Accepted for presentation at
Raleigh, North Carolina. October 2021. (4-page abstract
"Extending 1D Transport Using Neural Nets to GPUs"
(with P. Brantley). Accepted for presentation at
Tokyo, Japan. May 2020. (abstract only
"Transitioning the Scientific Software Toolchain to Clang/LLVM"
(with S. Dawson, R. Bleile, P. Brantley, M. McKinley, M. O'Brien, D. Richards). Accepted for presentation at
Paris, France. April 2020.
"Status of LLNL Monte Carlo Transport Codes on Sierra GPUs"
(with M. McKinley, R. Bleile, P. Brantley, S. Dawson, M. O'Brien, D. Richards). In
Proceedings of M&C 2019,
2160-2165. Portland, Oregon. August 2019.
"1D Transport Using Neural Nets, SN, and MC."
Proceedings of M&C 2019,
876-885. Portland, Oregon. August 2019.
"Porting the Opacity Client Library to a CPU-GPU Cluster Using
(with J. Kimko, R. Haque, and L. Grinberg). In
Proceedings of SC17.
Denver, Colorado. November 2017.
"Introduction to Monte Carlo."
LLNL's Computation Summer Student Seminar Series,
June, 2017 and
W&M Math Department Colloquium Series,
"LLNL Monte Carlo Transport Research Efforts for Advanced Computing Architectures"
(with P. Brantley, R. Bleile, S. Dawson, N. Gentile, M. McKinley, M. O'Brien, D. Richards, D.
Stevens, J. Walsh, and H. Childs). In
Proceedings of M&C 2017.
Jeju, Korea. April 2017.
"Optimizing Application I/O by Leveraging the
Storage Hierarchy Using the Scalable Checkpoint Restart Library
with a Monte Carlo Particle Transport Application on the Trinity
Advanced Computing System"
(with G. Becker, P. Brantley, S. Dawson, K. Mohror, A. Moody,
and M. O'Brien). In
Proceedings of SC16.
Salt Lake City, Utah. November 2016.
"Creating a Framework for Systematic Benchmarking of
High Performance Computing Systems."
Proceedings of SC14.
New Orleans, Louisiana. November 2014.
SNA+MC is the Joint International Conference on Supercomputing in Nuclear Applications + Monte Carlo.
EuroLLVM is the European LLVM Developers' Meeting.
M&C is the International Conference on Mathematics and Computational Methods applied to Nuclear Science and Engineering.
SC is the International Conference for High Performance Computing, Networking, Storage, and Analysis.