More Bang for Your Buck: Improved use of GPU Nodes for GROMACS 2018

TL;DR

This paper benchmarks various hardware configurations to identify cost-effective GPU-accelerated nodes for GROMACS 2018, demonstrating improved performance-to-price ratios and upgrade strategies leveraging GPUs.

Contribution

It provides updated performance benchmarks and cost analyses for GPU versus CPU nodes specifically optimized for GROMACS 2018 and later versions.

Findings

GPU nodes have higher performance-to-price ratios than CPU nodes.

Optimal GROMACS 2018 performance is achieved with GPU-accelerated hardware.

Upgrading old nodes with new GPUs is a cost-effective way to enhance performance.

Abstract

We identify hardware that is optimal to produce molecular dynamics trajectories on Linux compute clusters with the GROMACS 2018 simulation package. Therefore, we benchmark the GROMACS performance on a diverse set of compute nodes and relate it to the costs of the nodes, which may include their lifetime costs for energy and cooling. In agreement with our earlier investigation using GROMACS 4.6 on hardware of 2014, the performance to price ratio of consumer GPU nodes is considerably higher than that of CPU nodes. However, with GROMACS 2018, the optimal CPU to GPU processing power balance has shifted even more towards the GPU. Hence, nodes optimized for GROMACS 2018 and later versions enable a significantly higher performance to price ratio than nodes optimized for older GROMACS versions. Moreover, the shift towards GPU processing allows to cheaply upgrade old nodes with recent GPUs, yielding essentially the same performance as comparable brand-new hardware.