Jeremy L Thompson¶
I am a research software engineer, applied mathematician, and programming and mathematics educator. My experience includes statistical analysis for the U.S. Air Force and performance portable software development as part of the Department of Energy Center for Efficient Exascale Discretizations. I have professional experience in C, Rust, Python, C++, CUDA, Julia, Fortran, and R, as well as experience teaching several other languages. I have taught at the U.S. Air Force Academy, University of Colorado Boulder, and online at freeCodeCamp.
libCEED provides fast algebra for element-based discretizations, designed for performance portability, run-time flexibility, and clean embedding in higher level libraries and applications. It offers a C99 interface as well as bindings for Fortran, Python, Julia, and Rust. While our focus is on high-order finite elements, the approach is mostly algebraic and thus applicable to other discretizations in factored form.
Local Fourier Analysis is a tool commonly used in the analysis of multigrid and multilevel algorithms for solving partial differential equations via finite element or finite difference methods. This analysis can be used to predict convergence rates and optimize parameters in multilevel methods and preconditioners. This package provides a toolkit for analyzing the performance of preconditioners for arbitrary, user provided weak forms of partial differential equations.