Parallel Computing of Discrete Element Method on GPU

TL;DR

This paper explores GPU acceleration for the Discrete Element Method, demonstrating significant speed improvements with practical contact models, though with some trade-offs in computational efficiency.

Contribution

It introduces a modified contact force model for GPU-based DEM that balances realism and computational speed.

Findings

GPU outperforms CPU in DEM simulations

Practical contact model achieves 6x speedup on GPU

Simple model is 7x faster than practical model on GPU

Abstract

We investigate applicability of GPU to DEM. NVIDIA's code obtained superior performance than CPU in computational time. A model of contact forces in NVIDIA's code is too simple for practical use. We modify this model by replacing it with the practical model. The simulation shows that the practical model obtains the computing speed 6 times faster than the practical one on CPU while 7 times slower than the simple one on GPU. The result are analyzed.