January 13, 2023
Journal Article

Linear solvers for power grid optimization problems: a review of GPU-accelerated linear solvers

Abstract

The linear equations that arise in interior methods for constrained optimization are sparse symmetric indefinite and become extremely ill-conditioned as the interior method converges. These linear systems present a challenge for existing solver frameworks based on sparse LU or \LDLT{} decompositions. We benchmark five well known direct linear solver packages using matrices extracted from power grid optimization problems. The achieved solution accuracy varies greatly among the packages. None of the tested packages delivers significant GPU acceleration for our test cases.

Published: January 13, 2023

Citation

Swirydowicz K., E. Darve, W. Jones, J. Maack, S. Regev, M.A. Saunders, and S. Thomas, et al. 2022. Linear solvers for power grid optimization problems: a review of GPU-accelerated linear solvers. Parallel Computing 111. PNNL-SA-157570. doi:10.1016/j.parco.2021.102870