Accelerating the problem of microrheology in colloidal systems on a GPU

Complex fluids are characterized with both solid and fluid properties by their elasticity and viscosity, or rheological behaviour. The flow of complex fluids is a focus of interest in a very wide range of applications in biophysics and soft matter, such as problems involving live cells, processing of plastic, glass, paints, foods, oil recovery and so on. Recently, microrheology has been developed as an accurate technique to obtain rheological properties in soft matter from the microscopic motion of colloidal tracers used as probes, either freely diffusing in the host medium (passive), or subjected to external forces (active). A drawback for the models that simulate these techniques is their high computational cost. Therefore, the use of high performance computing is mandatory to develop the microrheology models. In this work, a microrheology model based on simulations of a tracer in a bath of Brownian quasi-hard spheres is proposed and parallelized. The analysis of the results of the evaluation using different sizes of the problem shows that GPU programming is very appropriate to accelerate these kinds of models.