Polymer Field-Theory Simulations on Graphics Processing Units

TL;DR

This paper presents the first GPU implementation of polymer field-theory simulations using CUDA, achieving significant speedups over CPU calculations for equilibrium property evaluations in polymer systems.

Contribution

The authors develop and demonstrate a GPU-based framework for polymer field-theoretic simulations, enabling faster and more efficient calculations compared to traditional CPU methods.

Findings

GPU implementation achieves up to 30x speedup for mean-field calculations.

Fully fluctuating simulations run up to 60x faster on GPU than CPU.

GPU approach outperforms MPI CPU clusters in communication overhead and speed.

Abstract

We report the first CUDA graphics-processing-unit (GPU) implementation of the polymer field-theoretic simulation framework for determining fully fluctuating expectation values of equilibrium properties for periodic and select aperiodic polymer systems. Our implementation is suitable both for self-consistent field theory (mean-field) solutions of the field equations, and for fully fluctuating simulations using the complex Langevin approach. Running on NVIDIA Tesla T20 series GPUs, we find double-precision speedups of up to 30x compared to single-core serial calculations on a recent reference CPU, while single-precision calculations proceed up to 60x faster than those on the single CPU core. Due to intensive communications overhead, an MPI implementation running on 64 CPU cores remains two times slower than a single GPU.