PhotoNs-GPU:A GPU accelerated cosmological simulation code

TL;DR

This paper introduces PhotoNs-GPU, a GPU-accelerated cosmological simulation code that significantly improves efficiency over CPU versions by optimizing GPU utilization and employing a novel interpolated method for truncated gravity calculations.

Contribution

The paper presents a new GPU-based cosmological simulation code with optimized algorithms and interpolation techniques, achieving high performance and efficiency improvements over existing CPU codes.

Findings

Single precision runs are twice as fast as double precision.

GPU implementation achieves up to 48% of theoretical peak performance.

The code maintains unbiased small noise in the power spectrum.

Abstract

We present a GPU-accelerated cosmological simulation code, PhotoNs-GPU, based on algorithm of Particle Mesh Fast Multipole Method (PM-FMM), and focus on the GPU utilization and optimization. A proper interpolated method for truncated gravity is introduced to speed up the special functions in kernels. We verify the GPU code in mixed precision and different levels of interpolated method on GPU. A run with single precision is roughly two times faster that double precision for current practical cosmological simulations. But it could induce a unbiased small noise in power spectrum. Comparing with the CPU version of PhotoNs and Gadget-2, the efficiency of new code is significantly improved. Activated all the optimizations on the memory access, kernel functions and concurrency management, the peak performance of our test runs achieves 48% of the theoretical speed and the average performance approaches to 35% on GPU.