Speedup of UEDGE Parameter Scans Using Machine‐Learning Optimized OpenMP Parallelization and a Continuation Solver
Contributions to Plasma Physics · 2026-11 · 66 (10)
DOI: 10.1002/ctpp.70135 ↗Kurz gefasst
Fachartikel in der Zeitschrift Contributions to Plasma Physics, online erschienen am 19.05.2026. Verfasst von A. Holm und 4 weiteren Autor:innen. Die Arbeit nennt 17 Literaturangaben. Gefördert von: U.S. Department of Energy, Lawrence Livermore National Laboratory.
Abstract (Originalsprache)
This work presents the OpenMP parallelization of the preconditioning Jacobian assembly and right‐hand side residual evaluation in UEDGE. A continuation algorithm, utilizing the internal NKSOL implicit Jacobian‐Free Newton‐Krylov solver to efficiently scan physical parameters, is also presented. The implemented parallelization reduces the computational time for a benchmark scan run on 32 threads by compared to the serial version when using trained random forest regression models to identify the optimal decomposition of the system of equations. Random forest regression models applied to the UEDGE time‐dependent and continuation solver algorithms did not yield meaningful improvement in computational performance. A benchmark DIII‐D gas injection rate scan in the 0.35–0.75 kA interval, performed on a test cluster using the parallelized code and continuation solver, produced 1066 steady‐state solutions with a 22 s average wall‐clock computational time per steady‐state solution.
Förderung
- U.S. Department of Energy
- Lawrence Livermore National Laboratory
Zitieren
APA
Holm, A., Guterl, J., Rognlien, T. D., Meyer, W. H., & Brown, P. N. (2026). Speedup of UEDGE Parameter Scans Using Machine‐Learning Optimized OpenMP Parallelization and a Continuation Solver. Contributions to Plasma Physics, 66(10). https://doi.org/10.1002/ctpp.70135
Vancouver
Holm A, Guterl J, Rognlien TD, Meyer WH, Brown PN. Speedup of UEDGE Parameter Scans Using Machine‐Learning Optimized OpenMP Parallelization and a Continuation Solver. Contributions to Plasma Physics. 2026;66(10). doi:10.1002/ctpp.70135
BibTeX
@article{Holm2026,
author = {Holm, A. and Guterl, J. and Rognlien, T. D. and Meyer, W. H. and Brown, P. N.},
title = {Speedup of UEDGE Parameter Scans Using Machine‐Learning Optimized OpenMP Parallelization and a Continuation Solver},
journal = {Contributions to Plasma Physics},
year = {2026},
volume = {66},
number = {10},
doi = {10.1002/ctpp.70135}
}
EUVIMED