Neutron particle transport 3D method of characteristic Multi GPU platform Parallel Computing

2025-03-22

AI Summary

Traditional Method of Characteristics (MOC) for large-scale 3D neutron transport simulations is severely constrained by GPU memory capacity and computational efficiency, hindering high-fidelity real-time simulation. Method: This paper proposes the first GPU-native real-time characteristic ray generation and preloading mechanism, integrated with dynamic load balancing and multi-GPU cooperative scheduling. Implemented within the CUDA framework, our approach employs memory-aware parallel MOC solving via fine-grained task decomposition, GPU memory reuse optimization, and communication overhead suppression. Contribution/Results: Our method overcomes the single-GPU memory bottleneck, enabling ultra-large-scale spatial discretization. Experiments demonstrate a 300–400× speedup over CPU-based serial MOC on multi-GPU platforms, with no loss in numerical accuracy. To the best of our knowledge, this is the first work to achieve high-resolution, real-time 3D MOC transport simulation.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Web Mining and Content Analysis: Web data generation and simulation
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

Three-dimensional neutron transport calculations using the Method of Characteristics (MOC) are highly regarded for their exceptional computational efficiency, precision, and stability. Nevertheless, when dealing with extensive-scale computations, the computational demands are substantial, leading to prolonged computation times. To address this challenge while considering GPU memory limitations, this study transplants the real-time generation and characteristic line computation techniques onto the GPU platform. Empirical evidence emphasizes that the GPU-optimized approach maintains a heightened level of precision in computation results and produces a significant acceleration effect. Furthermore, to fully harness the computational capabilities of GPUs, a dual approach involving characteristic line preloading and load balancing mechanisms is adopted, further enhancing computational efficiency. The resulting increase in computational efficiency, compared to traditional methods, reaches an impressive 300 to 400-fold improvement.

Problem

• Enhance 3D neutron transport computation speed
• Overcome GPU memory limitations efficiently
• Achieve 300-400x faster MOC calculations

Innovation

• 3D neutron transport using Method of Characteristics
• GPU-optimized real-time generation and computation
• Preloading and load balancing for efficiency