EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Adaptive mixed-precision Monte Carlo integration on GPUs

Ferhat Onur Özgan, Berke Kabasakal, Fahreddin Şükrü Torun

Computing · 2026-04 · 108 (4)

DOI: 10.1007/s00607-026-01656-7 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computing, online erschienen am 06.04.2026. Verfasst von Ferhat Onur Özgan und 2 weiteren Autor:innen. Die Arbeit nennt 15 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ankara Yıldırım Beyazıt University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Monte Carlo integration (MCI) is widely used for evaluating high-dimensional or non-analytic functions, but its large number of function evaluations can make it computationally demanding. As modern scientific applications increasingly rely on GPU acceleration, balancing numerical accuracy and performance has become a key challenge. To address this, we present a GPU-accelerated mixed-precision MCI framework that adaptively selects precision based on local numerical behavior using heuristics derived from gradient, variance, and average value analysis. Two precision allocation strategies are explored: term-wise and region-wise, both implemented with CUDA batch processing to maximize GPU efficiency. Experimental results on representative test functions of varying dimensionality demonstrate speedups of up to 4.9 $$\times $$ compared to full double precision, while maintaining controlled relative error. The proposed framework achieves a practical balance between computational efficiency and numerical reliability. It allows integration tasks on GPUs to run accurately and to adapt their precision automatically for higher performance.

Förderung

  • Ankara Yıldırım Beyazıt University

Zitieren

APA

Özgan, F. O., Kabasakal, B., & Torun, F. Ş. (2026). Adaptive mixed-precision Monte Carlo integration on GPUs. Computing, 108(4). https://doi.org/10.1007/s00607-026-01656-7

Vancouver

Özgan FO, Kabasakal B, Torun FŞ. Adaptive mixed-precision Monte Carlo integration on GPUs. Computing. 2026;108(4). doi:10.1007/s00607-026-01656-7

BibTeX
@article{zgan2026,
  author = {Özgan, Ferhat Onur and Kabasakal, Berke and Torun, Fahreddin Şükrü},
  title = {Adaptive mixed-precision Monte Carlo integration on GPUs},
  journal = {Computing},
  year = {2026},
  volume = {108},
  number = {4},
  doi = {10.1007/s00607-026-01656-7}
}
Kostenlos starten →