What are the fastest and most efficient ways to run Molecular Dynamics (MD) simulations?
A poor choice of job submission or simulation parameters leads to poor performance and a waste of computing resources. A common mistake is requesting too many CPUs or GPUs to run a simulation as fast as possible. However, the fastest simulations may be inefficient due to poor parallel scaling. As a result, such simulations use excessive resources and negatively impact priority, resulting in longer queue times and reduced productivity.
Finding optimal MD engines and submission parameters in a complex High Performance Computing (HPC) environment with heterogeneous hardware poses a daunting and time-consuming challenge. We recently launched an interactive molecular dynamics performance guide to simplify this process.
This web portal presents performance metrics of all installed MD software packages. It includes a searchable benchmarks database with simulation details such as hardware descriptions, SLURM submission scripts and input files for simulations. Using the plots that show the speed of all tested MD executables and simulation cost, it is easy to predict how long it will take for a simulation to complete and how much resources will be needed.
Currently, the database contains over 130 benchmarks for four major MD software packages: AMBER, GROMACS, NAMD and OpenMM. Efforts to test all possible hardware/software combinations are underway.
We can help you with your molecular dynamics simulations. Just contact us at support@ace-net.ca.
