|
Steve Plimpton |
 |
|
I retired in October 2022 from Sandia National Laboratories, a
US Department of Energy lab, where I worked as a computational
scientist in its Center for Computing Research.
I'm still involved in some technical work, particularly for the open source codes listed below.
You can reach me with a question or idea at this email address --- sjplimp at gmail.com --- though I may not respond immediately :)
My research at Sandia involved implementing and using scientific
simulations designed for parallel supercomputers, as well as the
creation of efficient parallel algorithms. Many of the models I
worked on were particle based; some used structured grids or finite
elements. Most of the codes I was a developer for were for materials
modeling, some for biology and informatics problems.
- LAMMPS = classical molecular dynamics
- SPARTA = Direct Simulation Monte Carlo (DSMC)
- SPPARKS = kinetic and Metropolis Monte Carlo
- ChemCell = spatial reaction/diffusion in biological cells
- CSlib = client/server messaging library
- fftMPI = parallel 3d/2d FFT library
- MapReduce-MPI = MapReduce library on top of MPI
- PHISH = stream processing library on top of MPI
- FireHose = stream processing benchmarks
- CrossSim = simulator for modeling resisitve memory (analog) devices for neuromorphic computing
- Pizza.py = Python-based pre- and post-processing tools for LAMMPS, SPPARKS, SPARTA, ChemCell
Publications, see more info at Google Scholar
University course in parallel computing
Humorous quotations
Current Interests
Molecular Dynamics (MD): LAMMPS, ParaDyn, parallel algorithms, simulations (papers, pictures, movies)
Direct Simulation Monte Carlo - low-density reacting flows via particles, SPARTA and Icarus
Kinetic Monte Carlo - KMC and Metropolis MC simulator SPPARKS
Parallel FFTs - parallel distributed 3d/2d FFTs
Pizza.py Toolkit - Python-based tools for simulation setup, analysis, plotting, viz
Past Projects
Informatics - data-intensive computing via a MapReduce-MPI library, and a stream-processing PHISH library
High Performance Computing (HPC) - neuro-inspired computing, performance comparisons of parallel architectures
Biological Cell Modeling - particle-based reaction/diffusion simulator ChemCell
Parallel Algorithms: grid transfer, dynamic load-balancing, tinkertoy parallel programming, matrix-vector multiply and many-body algorithms, HPCC RandomAccess (GUPS) benchmark
Gene Finding - parallelization of the Genehunter genetic linkage analysis program
Radiation Transport - Boltzmann equation for thermal and X-ray radiation
ChISELS - surface evolution model of semiconductor processing for micromachine devices (MEMS)
QuickSilver - particle-in-cell (PIC) relativistic electromagnetics
PRONTO - finite element transient dynamics (crash & boom) with contact detection
NIMROD - plasma simulation of tokamaks
Electronic Structure - quantum mechanics of solids
Image Processing - synthetic aperture radar (SAR) analysis
Grain Growth - Potts model on 2-d and 3-d lattices
Electron Microscopy - Monte Carlo trajectory simulation
Parallel Computing - general overviews