Targeting GPUs with OpenMP Directives on Summit: A Simple and Effective Fortran Experience
arXiv:1812.07977 · doi:10.1016/j.parco.2019.102544
Abstract
We use OpenMP to target hardware accelerators (GPUs) on Summit, a newly deployed supercomputer at the Oak Ridge Leadership Computing Facility (OLCF), demonstrating simplified access to GPU devices for users of our astrophysics code GenASiS and useful speedup on a sample fluid dynamics problem. We modify our workhorse class for data storage to include members and methods that significantly streamline the persistent allocation of and association to GPU memory. Users offload computational kernels with OpenMP target directives that are rather similar to constructs already familiar from multi-core parallelization. In this initial example we ask, "With a given number of Summit nodes, how fast can we compute with and without GPUs?", and find total wall time speedups of . We also find reasonable weak scaling up to 8000 GPUs (1334 Summit nodes). We make available the source code from this work at https://github.com/GenASiS/GenASiS_Basics.
arXiv commentVersion accepted by Parallel Computing: Systems and Applications
References in corpus (7)
- Turbulent Magnetic Field Amplification from Spiral SASI Modes: Implications for Core-Collapse Supernovae and Proto-Neutron Star Magnetization
- Generation of Magnetic Fields by the Stationary Accretion Shock Instability
- Stochasticity and efficiency of simplified core-collapse supernova explosions
- GenASiS: General Astrophysical Simulation System. I. Refinable Mesh and Nonrelativistic Hydrodynamics
- GenASiS Basics: Object-oriented utilitarian functionality for large-scale physics simulations (Version 4)
- Turbulence and magnetic field amplification from spiral SASI modes in core-collapse supernovae
- GenASiS Mathematics: Object-oriented manifolds, operations, and solvers for large-scale physics simulations