GPU accelerated Hybrid Tree Algorithm for Collision-less N-body Simulations

TL;DR

This paper introduces a GPU-accelerated hybrid tree algorithm for collision-less N-body simulations that significantly reduces computation and communication costs by splitting forces into hard and soft components and applying tailored integration methods.

Contribution

The paper presents a novel hybrid tree algorithm that efficiently splits force calculations and implements it on GPU clusters to improve performance in N-body simulations.

Findings

Reduced communication cost to 40% of normal tree algorithm

Lowered total execution time to 80% of normal tree algorithm

Potential to reduce execution time to about 70% with more processes

Abstract

We propose a hybrid tree algorithm for reducing calculation and communication cost of collision-less N-body simulations. The concept of our algorithm is that we split interaction force into two parts: hard-force from neighbor particles and soft-force from distant particles, and applying different time integration for the forces. For hard-force calculation, we can efficiently reduce the calculation and communication cost of the parallel tree code because we only need data of neighbor particles for this part. We implement the algorithm on GPU clusters to accelerate force calculation for both hard and soft force. As the result of implementing the algorithm on GPU clusters, we were able to reduce the communication cost and the total execution time to 40% and 80% of that of a normal tree algorithm, respectively. In addition, the reduction factor relative the normal tree algorithm is smaller for large number of processes, and we expect that the execution time can be ultimately reduced down to about 70% of the normal tree algorithm.