Task-parallelism in SWIFT for heterogeneous compute architectures

Le résumé fourni par la source

ABSTRACT This paper highlights first steps towards enabling graphics processing unit (GPU) acceleration of the task-parallel smoothed particle hydrodynamics (SPH) solver SWIFT. Novel combinations of algorithms are presented, enabling SWIFT to function as a truly heterogeneous software leveraging task-parallelism on CPUs for memory-bound computations concurrently with GPUs for compute-bound computations while minimizing the effects of CPU–GPU communication latency. The proposed algorithms are validated in extensive testing. The GPU acceleration methodology is shown to deliver up to $\sim$3.5 and $\sim 7.5$ speedups for the offloaded computations when including and excluding the time required to prepare and post-process data transfers on the CPU side, respectively. The overall performance of the GPU-accelerated hydrodynamic solver for a full simulation on a single Grace Hopper superchip is 1.8 times faster compared to the superchip’s fully parallelized CPU capabilities. This constitutes an improvement from $\sim$8 million particle updates/s for the full CPU-only baseline ($\sim$115 000 updates per CPU core) to $\sim$15 million updates/s for the GPU-accelerated SPH solver. Moreover, it displays near-perfect strong scaling on 4 Grace Hopper nodes. The GPU-acceleration is also demonstrated to give a 29 per cent improvement in energy efficiency in comparison to CPU-only baselines. Finally, inter-influential bottlenecks in the prototype solver presented in this work are identified: a significant amount of time (up to 80 per cent) of a GPU-offloading cycle is spent on preparing and post-processing particle data on the CPU for the transfer to and from the GPU, respectively. Approaches are suggested to minimize their effects and maximize the solver’s performance in our future work.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.