Computational acceleration strategies for large-scale energy system optimization: a comparative study of GPU-accelerated and distributed-memory solvers
under review
- Energy system optimization models are increasing in scope and resolution, integrating detailed technology representations, sector coupling, and multiple scenarios reflecting uncertainty in future energy demand. These advances yield large and challenging linear programs whose efficient solution remains a bottleneck in practical energy system analyses. For a long time, the standard way to address such problems has relied on shared-memory interior-point methods (IPM), which combine robustness and accuracy but face scalability limits as model instance size grows. Specialized solver architectures are beginning to change this picture. Two promising directions have emerged: (i) GPU-accelerated first-order methods (FOM) such as the primal-dual linear programming approach; and (ii) distributed-memory IPM, exemplified by the open-source solver PIPS-IPM++, which can exploit block structure that arises in many energy system models and achieves high parallelism on highperformance computing systems. These developments open new opportunities for large-scale optimization in energy system analysis. This paper presents a computational study comparing these solver classes on a diverse test set of largescale linear programs arising from energy system analysis, including scenario-based formulations derived from stochastic programming. We investigate how parallelization strategies affect solution time and numerical accuracy. The results illustrate that distributed-memory IPM can leverage problem structure to deliver substantial speed-ups on specific problems with block-angular structures. GPU-accelerated FOMs demonstrate strong scalability but may yield solutions with higher relative infeasibilities, which, depending on the use case and model uncertainty, can still be acceptable. Overall, our findings indicate that recent algorithmic and hardware advances substantially broaden the computational toolbox available to the energy system optimization community. Each solver class exhibits distinct advantages: shared-memory IPMs remain a powerful tool for reliably obtaining high-accuracy solutions; distributed-memory IPMs can extend scalability to hundreds of cores for certain structured models, enabling faster time-to-solution; and GPU-based FOM can deliver fast solutions when such lower accuracy levels are appropriate. Together, they help make high-resolution, multi-scenario energy system optimization models tractable across a broader range of problem sizes and computing environments.
| Author: | Janina ZittelORCiD, Annika Buchholz, Michael Bussieck, Frederik Fiand, Thorsten Koch, Niels Lindner, Lukas Mehl, Manuel Wetzel |
|---|---|
| Document Type: | In Proceedings |
| Parent Title (English): | ECOS 2026 |
| Year of first publication: | 2026 |
| ArXiv Id: | http://arxiv.org/abs/2605.04605 |

