GROMACS

What is GROMACS

GROMACS is a versatile package for performing classical molecular dynamics simulations. Such simulations provide a three-dimensional description of a molecular system and its time evolution by integrating Newton’s equations of motion for problems containing hundreds to millions of particles. GROMACS is primarily designed for modelling biochemical molecules like proteins, lipids and nucleic acids characterised by large numbers of complex interactions. The size and time scales of biological systems call for the use of classical mechanics in conjunction with physically validated force fields, which describe all the parameters needed to run realistic simulations. Currently, many research groups are also using GROMACS to study non-biological systems, like polymers or fluid dynamics problems.

GROMACS is one of the most widely used high-performance computing (HPC) applications and is one of the fastest molecular dynamics codes. In addition to a large number of algorithms commonly used for molecular dynamics studies (see the GROMACS documentation page ), GROMACS also features several novel methods such as heterogeneous parallelisation and GPU acceleration, the accelerated weight histogram (AWH) method to accelerate sampling and to calculate free energy for alchemical transformation , hybrid Quantum-Classical simulations (QM/MM) with CP2K interface , and computational electrophysiology .

GROMACS development partners

GROMACS is a community-driven project, built on contributions from all over the world. At present, the development is being led by researchers from the KTH Royal Institute of Technology and the Science for Life Laboratory in Stockholm.

Current GROMACS development at PDC

GROMACS development at PDC is oriented towards improving molecular dynamics acceleration by implementing algorithms and optimising code for graphics processing units (GPUs) and heterogeneous HPC platforms (for example, by modernising communication algorithms and tuning of computation kernels) or by designing new solutions to reduce the overhead in memory transfers and computation (for example, by employing GPU-initiated communication). PDC is also working on porting GROMACS to current and new architectures and developing hardware/software co-design strategies. Current projects at PDC include modernised communication algorithms, SYCL porting for new accelerator architectures and implementation of fixed precision for coordinates, together with designing new strategies that promote the use of HPC resources for life and materials science problems. PDC also focuses on developing software sustainability aspects for scientific codes such as GROMACS.

A part of the work is supported by the European projects BioExcel, GANANA and DARE.