414 Cloud-based molecular simulation for industrial chemical engineering applications

The Chair of Thermodynamics and Energy Technology (ThEt) at the University of Paderborn develops the molecular simulation tool ms2 that is aimed at thermophysical properties of condensed fluids. This simulation software is currently used by ThEt for commercial applications on its workstation computing systems. Based on molecular force field methods, nowadays powerful predictive methods exist that yield thermophysical data which are the basis for the design and optimization of chemical engineering processes. Typically, the chemical industry measures the required data experimentally, however, if they are needed for hazardous substances (explosive, toxic or mutagenic), the associated costs may be prohibitive. In such cases, molecular dynamics (MD) and Monte Carlo (MC) simulations on the basis of optimized force fields are very attractive alternative routes. To carry out such predictions with molecular simulation, typically a larger number of simulation runs is necessary. The advantage of cloud-based simulations for this experiment is, that the individual simulations are mutually independent and that therefore, the whole experiment features very loose coupling. Therefore, this experiment can make use of cloud infrastructure, which is a lot less cost intensive than conventional HPC resources. Furthermore, the experiment will feature dynamic adaptivity to the obtained results – zooming in on relevant areas of the phase space. Here, the dynamic allocation possibilities in clouds will be very powerful.