Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
January 26, 2026Nuclear Science and Technology Open ResearchOpen Access

Development and verification of Hornet – A CUDA-optimized, Monte Carlo-agnostic, continuous-energy cross-section processing code

View Full Paper
Ask AI
Bookmark
Share

Authors

MAMuhammad Rizwan AliMAMurat Serdar AygulDLDeokjung Lee

Discussion

Loading...

Member takes

Overview

Hornet demonstrates efficient nuclear data processing in various Monte Carlo applications, indicating its robust versatility.

Key Points

  • Hornet aims to provide a GPU-optimized solution for processing neutron cross-sections and sampling thermal scattering laws.
  • Implemented using CUDA C++ for GPU parallelism
  • Verification through integration with GREAPMC across standard benchmarks
  • Assessed performance and precision against a CPU-based Monte Carlo code
  • Matched CPU-based code in accuracy with RMS power differences below 0.1%
  • Showed improved tracking rates compared to traditional Monte Carlo simulations
  • Successfully modeled continuous-energy and thermal scattering effects on GPU hardware

Cite This Study

Ali et al. (2026) studied this question.

synapsesocial.com/papers/6977032e722626c4468e833dhttps://doi.org/10.12688/nuclscitechnolopenres.17676.2
View Full Paper
Ask AI
Bookmark
Share