Nihar
Kodkani
Anvil REU Molecular Chemistry Gateway: Developing Open OnDemand Applications for Purdue's Anvil Cluster Mathematical/Computation Sciences
Abstract profile. Full document pending author claim.
Authors:
Nihar Kodkani
Date Created:
Not specified
Course Title:
Professor:
Not specified
About Paper:
Computational chemists utilize packages like GROMACS for molecular dynamics simulations and VASP for performing quantum-mechanics simulations. Running these calculations and simulations is a processor- intensive task, thus benefiting significantly from high-performance computing (HPC) clusters. Although these molecular simulation packages are available on the Anvil super computing cluster from the command line, having an interactive web gateway will help inexperienced users. This project aims to integrate GROMACS and VASP as interactive applications within the Open OnDemand (OOD) portal. We focused first on implementing GROMACS, as it is a widely used package in chemistry and biological domains. Before implementing, we researched parameters for the various simulations. After extensive consultations with computational chemists and GROMACS users, a standard set of parameters was determined. To make job submission easier for new users, the application provided an optional Python script to automatically generate .mdp files (simulation configuration files) given user-provided parameters. The GROMACS OOD application contains an option to generate simulation movies. This was created using a combination of VMD and FFMPEG (visualization and video software). We additionally implemented the VASP application, another popular application in the quantum chemistry community. The initial development process was similar to GROMACS, but the parameters varied depending on the type of simulations. In addition, the application had pre-generated configuration files like GROMACS. Implementing the gateway makes it much easier for Anvil users to apply Gromacs and VASP for their research, allowing computational chemists and researchers to focus on their research without needing to navigate the complexities of Linux commands. Keywords: Molecular Dynamics; Computational Chemistry; Anvil; HPC; Gromacs
Source:
Purdue University / 2024
Topics:
No topics listed
Co-authors:
Nihar Kodkani