High Performance Computing in Multiscale Problems of Gas Dynamics

The work is devoted to the organization of high performancecomputing in the solution of multiscale problems of gas dynamicsrelevant for the implementation of modern nanotechnologies. Thebase of the presented computing technology is a multiscale two-levelapproach that combines calculations at macro- and microlevels. Theapproach makes it possible to study micro- and nanoflows of agaseous medium under conditions of complex geometry of technicalsystems used in the production cycle in order to obtain newnanomaterials and nanocoatings. Within the framework of theapproach a system of the quasigasdynamic (QGD) equations and a system of the molecular dynamics (MD) equations are considered as two basic mathematical models. These models are aggregated using themethod of splitting by physical processes and scales. The QGDsystem is solved by the finite volume method on grids of arbitrarytype. The MD equations are solved according to the Verletintegration. In view of the complexity of the problem a highperformance computing is used for realization of the approach.Parallel implementation of the approach is based on the methods ofdomain decomposition and functional parallelism and is orientedtowards the use of computer systems with hybrid architecture. Theimplementation uses MPI, OpenMP and CUDA technologies. Testing ofthe developed approach and parallel tool was performed using theexample of the problem of spraying the nanoparticles on asubstrate. Numerical experiments confirm the effectiveness of thedeveloped computing technology