Parallel distributed computing in modeling of the nanomaterials production technologies

Simulation of physical and chemical processes occurring in the nanomaterial production technologies is a computationally challenging problem, due to the great number of coupled processes, time and length scales to be taken into account. To solve such complex problems with a good level of detail in a reasonable time, not only high-performance computational resources are needed, but also special parallel methods should be developed to efficiently utilize parallel distributed resources of computer Grids. This paper presents a Virtual Reactor problem solving environment developed for simulation of plasma chemical deposition of nanomaterials, describes an adaptive workload balancing algorithm with user-level scheduling for parallel applications on heterogeneous Grid resources, and shows some performance results.