Open-source scientific computing
Software development is a core part of my research. Sustainability, reproducibility, and a low barrier to entry matter as much as the algorithms.
-
hIPPYlib
An extensible framework implementing state-of-the-art scalable algorithms for PDE-based deterministic and Bayesian inverse problems. Built on FEniCS/dolfinx for PDE discretization and PETSc for scalable linear algebra.
UT Austin · UC Merced — lead developers: U. Villa, N. Petra
-
hIPPYlib-MUQ
Couples hIPPYlib's scalable inversion machinery with MUQ's advanced MCMC samplers, for Bayesian inference with complex predictive models under uncertainty.
-
SOUPy
Stochastic PDE-constrained optimization under high-dimensional uncertainty, in Python. Built on hIPPYlib and FEniCS.
-
ParELAG
Upscaling and algebraic multigrid techniques for the efficient solution of algebraic systems arising from mixed finite element discretizations of saddle point problems.
Original author: U. Villa · Developers: A. Barker, T. Benson, C. Lee · PI: P. Vassilevski
-
MFEM
A lightweight, general, scalable C++ library for finite element methods, relying on HYPRE for fast parallel solvers and preconditioners.
LLNL — lead developers: T. Kolev, V. Dobrev