Article · Computation · March 20, 2024
Exploring Numba and CuPy for GPU-Accelerated Monte Carlo Radiation Transport
Tair Askar, Argyn Yergaliyev, Bekdaulet Shukirgaliyev, Ernazar B. Abdikamalov
Abstract
This paper examines the performance of two popular GPU programming platforms, Numba and CuPy, for Monte Carlo radiation transport calculations. We conducted tests involving random number generation and one-dimensional Monte Carlo radiation transport in plane-parallel geometry on three GPU cards: NVIDIA Tesla A100, Tesla V100, and GeForce RTX3080. We compared Numba and CuPy to each other and our CUDA C implementation. The results show that CUDA C, as expected, has the fastest performance and highest energy efficiency, while Numba offers comparable performance when data movement is minimal. While CuPy offers ease of implementation, it performs slower for compute-heavy tasks.
Related publications
- Evaluating multi-GPU computing capabilities of Numba and CuPyCluster Computing · 2025 · Citations: 1 · Same author: Tair Askar
- EFFICIENT COMPUTATIONAL APPROACHES TO GPU-BASED MONTE CARLO RADIATION TRANSPORTNazarbayev University Repository · 2025 · Same author: Tair Askar
- Evaluation of Pseudo-Random Number Generation on GPU CardsComputation · 2021 · Citations: 15 · Same author: Tair Askar
- Performance evaluation of GPU-based parallel sorting algorithmsPLoS ONE · 2026 · Citations: 4 · Shared topic: Parallel Computing and Optimization Techniques