Efficient GPU Implementation of Particle Interactions with Cutoff Radius and Few Particles per Cell

TL;DR

This paper introduces two GPU algorithms for efficiently computing particle interactions with cutoff radius in scenarios with few particles per cell, highlighting the X-pencil method's potential speedup under specific conditions.

Contribution

It proposes two novel GPU algorithms utilizing shared memory for particle interactions with cutoff, improving performance in specific cases.

Findings

X-pencil approach offers significant speedup in certain scenarios

Shared memory utilization enhances GPU performance for particle interactions

Comparison across three GPU models demonstrates method effectiveness

Abstract

This paper presents novel approaches to parallelizing particle interactions on a GPU when there are few particles per cell and the interactions are limited by a cutoff distance. The paper surveys classical algorithms and then introduces two alternatives that aim to utilize shared memory. The first approach copies the particles of a sub-box, while the second approach loads particles in a pencil along the X-axis. The different implementations are compared on three GPU models using Cuda and Hip. The results show that the X-pencil approach can provide a significant speedup but only in very specific cases.