Abelian-Higgs cosmic string evolution with multiple GPUs

TL;DR

This paper presents a scalable multi-GPU simulation code for Abelian-Higgs cosmic string networks, enabling large-scale, high-resolution cosmological simulations that significantly reduce computation time.

Contribution

The authors develop and validate a multi-GPU extension of their existing GPU-accelerated code, achieving high scalability and enabling large-scale simulations of cosmic string evolution.

Findings

Successful implementation of multi-GPU code with good scalability

Large $8192^3$ simulation run completed in 33.2 minutes using 4096 GPUs

Enhanced capability for accurate, large-scale cosmological defect simulations

Abstract

Topological defects form at cosmological phase transitions by the Kibble mechanism. Cosmic strings and superstrings can lead to particularly interesting astrophysical and cosmological consequences, but this study is is currently limited by the availability of accurate numerical simulations, which in turn is bottlenecked by hardware resources and computation time. Aiming to eliminate this bottleneck, in recent work we introduced and validated a GPU-accelerated evolution code for local Abelian-Higgs strings networks. While this leads to significant gains in speed, it is still limited by the physical memory available on a graphical accelerator. Here we report on a further step towards our main goal, by implementing and validating a multiple GPU extension of the earlier code, and further demonstrate its good scalability, both in terms of strong and weak scaling. A $8192^3$ production run, using $4096$ GPUs, runs in $33.2$ minutes of wall clock time on the Piz Daint supercomputer.