Parallel execution of a parameter sweep for molecular dynamics simulations in a hybrid GPU/CPU environment

MILLÁN, EMMANUEL NICOLÁS; CARLOS GARCÍA GARINO; EDUARDO M. BRINGA

Bahia Blanca

Congreso; Congreso Argentino de Ciencias de la Computación 2012; 2012

Universidad Nacional del Sur

Molecular Dynamics (MD) simulations can help to under- stand an immense number of phenomena at the nano and microscale. They often require the exploration of large parameter space, and a pos- sible parallelization strategy consists of sending different parameter sets to different processors. Here we present such approach using a hybrid environment of Graphic Processing Units (GPUs) and CPU cores. We take advantage of the software LAMMPS (lammps.sandia.gov), which is already prepared to run in a hybrid environment, in order to do an efficient parameter sweep. One example is presented in this work: the collision of two clusters is sampled over a multivariate space to obtain information on the resulting structural properties.