Parallel density matrix propagation in spin dynamics simulations

TL;DR

This paper introduces a parallel method for density matrix propagation in spin dynamics simulations that minimizes communication overhead, enabling efficient scaling on high-performance computing clusters.

Contribution

The paper proposes a novel parallel propagation approach that reduces communication costs, improving scalability in distributed spin dynamics simulations.

Findings

Effective scaling demonstrated on a 128-core cluster

Communication overhead significantly reduced

Method recasts simulation to require minimal inter-thread communication

Abstract

Several methods for density matrix propagation in distributed computing environments, such as clusters and graphics processing units, are proposed and evaluated. It is demonstrated that the large communication overhead associated with each propagation step (two-sided multiplication of the density matrix by an exponential propagator and its conjugate) may be avoided and the simulation recast in a form that requires virtually no inter-thread communication. Good scaling is demonstrated on a 128-core (16 nodes, 8 cores each) cluster.