Tackling Exascale Software Challenges in Molecular Dynamics Simulations with GROMACS

Abstract

GROMACS is a widely used package for biomolecular simulation, and over the last two decades it has evolved from small-scale efficiency to advanced heterogeneous acceleration and multi-level parallelism targeting some of the largest supercomputers in the world. Here, we describe some of the ways we have been able to realize this through the use of parallelization on all levels, combined with a constant focus on absolute performance. Release 4.6 of GROMACS uses SIMD acceleration on a wide range of architectures, GPU offloading acceleration, and both OpenMP and MPI parallelism within and between nodes, respectively. The recent work on acceleration made it necessary to revisit the fundamental algorithms of molecular simulation, including the concept of neighborsearching, and we discuss the present and future challenges we see for exascale simulation - in particular a very fine-grained task parallelism. We also discuss the software management, code peer review and continuous integration testing required for a project of this complexity.

EASC 2014 conference proceeding

References in corpus (3)

• A flexible algorithm for calculating pair interactions on SIMD architectures
• Best Practices for Scientific Computing
• A Tuned and Scalable Fast Multipole Method as a Preeminent Algorithm for Exascale Systems

Cited by in corpus (37)

• Computer Modeling of Halogen Bonds and Other $\sigma$-Hole Interactions
• On the molecular origins of the ferroelectric splay nematic phase
• GROMACS in the cloud: A global supercomputer to speed up alchemical drug design
• The good, the bad and the user in soft matter simulations
• On the Molecular Origin of the Cooperative Coil-to-globule Transition of Poly(N-isopropylacrylamide) in Water
• RUMD: A general purpose molecular dynamics package optimized to utilize GPU hardware down to a few thousand particles
• Aerosol-OT surfactant forms stable reverse micelles in aploar solvent in the absence of water
• Predictive Collective Variable Discovery with Deep Bayesian Models
• Pressure Profile Calculation with Mesh Ewald Methods
• Evidence of a low-temperature dynamical transition in concentrated microgels
• Molecular simulation of methane hydrate growth confined into a silica pore
• Adaptive multi-stage integrators for optimal energy conservation in molecular simulations
• Open Boundary Simulations of Proteins and Their Hydration Shells by Hamiltonian Adaptive Resolution Scheme
• Computing spatially resolved rotational hydration entropies from atomistic simulations
• Mapping Microstructural Dynamics up to the Nanosecond of the Conjugated Polymer P3HT in the Solid State
• Controlling ${\text{Li}}^{+}$ transport in ionic liquid electrolytes through salt content and anion asymmetry: A mechanistic understanding gained from molecular dynamics simulations
• Fluctuation relations to calculate protein redox potentials from molecular dynamics simulations
• Polymer electrolytes in strong external electric fields: Modification of structure and dynamics
• MiMiC: A High-Performance Framework for Multiscale Molecular Dynamics Simulations
• Performance Evaluation of the Zero-Multipole Summation Method in Modern Molecular Dynamics Software
• Scalable and fast heterogeneous molecular simulation with predictive parallelization schemes
• Application-Driven Exascale: The JUPITER Benchmark Suite
• GROMACS on AMD GPU-Based HPC Platforms: Using SYCL for Performance and Portability
• Breaking Down the Parallel Performance of GROMACS, a High-Performance Molecular Dynamics Software
• Competitive Adsorption in Polymer Nanocomposites: The Molecular Weight and End-Group Effect Revealed by SANS and MD Simulations
• Simulation-Based Inference with Approximately Correct Parameters via Maximum Entropy
• Reaction Path Averaging: Characterizing the Structural Response of the DNA Double Helix to Electron Transfer
• Rapid Conformational Analysis of Semi-Flexible Liquid Crystals
• MateriApps LIVE! and MateriApps Installer: Environment for starting and scaling up materials science simulations
• OpenMM-MiMiC Interface for Efficient and Flexible Multiscale Simulations
• ARENA: Asynchronous Reconfigurable Accelerator Ring to Enable Data-Centric Parallel Computing
• Designing a Prospective COVID-19 Therapeutic with Reinforcement Learning
• Dynamics of Polyalkylfluorene Conjugated Polymers: Insights from Neutron Spectroscopy and Molecular Dynamics Simulations
• Temperature Effect on Interactions of Oil Droplet with Water-wetted Shale Kerogen at Reservoir Temperatures
• tinyMD: A Portable and Scalable Implementation for Pairwise Interactions Simulations
• Phenylalanine modification in plasma-driven biocatalysis revealed by solvent accessibility and reactive dynamics in combination with protein mass spectrometry
• Global free energy landscapes as a smoothly joined collection of local maps