Kokkos-Based Implementation of MPCD on Heterogeneous Nodes

TL;DR

This paper presents a performance portable implementation of Multi-Particle Collision Dynamics (MPCD) using the Kokkos-based library Cabana, enabling efficient simulations on heterogeneous CPU-GPU systems and analyzing their interplay.

Contribution

It introduces a novel, portable MPCD implementation leveraging Cabana for heterogeneous systems, facilitating performance analysis across CPU and GPU usage.

Findings

Optimal performance on homogeneous GPU systems

Insights into CPU-GPU interplay for MPCD simulations

Discussion on heterogeneous application efficiency

Abstract

The Kokkos based library Cabana, which has been developed in the Co-design Center for Particle Applications (CoPA), is used for the implementation of Multi-Particle Collision Dynamics (MPCD), a particle-based description of hydrodynamic interactions. It allows a performance portable implementation, which has been used to study the interplay between CPU and GPU usage on a multi-node system. As a result, we see most advantages in a homogeneous GPU usage, but we also discuss the extent to heterogeneous applications, using both CPU and GPU concurrently.