A highly scalable particle tracking algorithm using partitioned global address space (PGAS) programming for extreme-scale turbulence simulations

TL;DR

This paper presents a scalable PGAS-based parallel particle tracking algorithm designed for extreme-scale turbulence simulations, significantly improving efficiency on petascale supercomputers by minimizing communication overhead.

Contribution

The work introduces a novel PGAS algorithm for particle tracking that reduces communication and enhances scalability in large turbulence simulations.

Findings

Achieved high scalability on 262,144 cores for an 8192^3 grid simulation.

Algorithm is orders of magnitude faster than previous methods.

Efficient one-sided communication implementation using Co-Array Fortran.

Abstract

A new parallel algorithm utilizing partitioned global address space (PGAS) programming model to achieve high scalability is reported for particle tracking in direct numerical simulations of turbulent flow. The work is motivated by the desire to obtain Lagrangian information necessary for the study of turbulent dispersion at the largest problem sizes feasible on current and next-generation multi-petaflop supercomputers. A large population of fluid particles is distributed among parallel processes dynamically, based on instantaneous particle positions such that all of the interpolation information needed for each particle is available either locally on its host process or neighboring processes holding adjacent sub-domains of the velocity field. With cubic splines as the preferred interpolation method, the new algorithm is designed to minimize the need for communication, by transferring between adjacent processes only those spline coefficients determined to be necessary for specific particles. This transfer is implemented very efficiently as a one-sided communication, using Co-Array Fortran (CAF) features which facilitate small data movements between different local partitions of a large global array. Detailed benchmarks are obtained on the Cray petascale supercomputer Blue Waters at the University of Illinois, Urbana-Champaign. For operations on the particles in a $8192^3$ simulation ($0.55$ trillion grid points) on $262,144$ Cray XE6 cores, the new algorithm is found to be orders of magnitude faster relative to a prior algorithm in which each particle is tracked by the same parallel process at all times. Improving support of PGAS models on major compilers suggests that this algorithm will be of wider applicability on most upcoming supercomputers.