On 46 large energy-system LPs, GPU first-order methods offer the largest speedups on the hardest instances at medium accuracy, distributed IPMs win on block-angular models, and shared-memory IPMs remain the accuracy baseline.
Available at https://github.com/NVIDIA/cuopt, [accessed 26.03.2026]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Computational acceleration strategies for large-scale energy system optimization: a comparative study of GPU-accelerated and distributed-memory solvers
On 46 large energy-system LPs, GPU first-order methods offer the largest speedups on the hardest instances at medium accuracy, distributed IPMs win on block-angular models, and shared-memory IPMs remain the accuracy baseline.