Ensemble simulations and optimization studies in the cloud - application examples for system design

In system design simulation-based methods play a key role. In the future new methods become important, which will be based on a large number of individual simulations and will lead to higher quality results. The designed systems will be statistically analyzed comprehensively and thus behave more robust under changing environmental conditions. Cloud computing offers scalable, on-demand shortterm usable computing power. This will be an important basis for processing resource-intensive, simulation-based design tasks in a time frame that is acceptable for the engineer. Based on a generic Split-Map-Reduce approach with Grid-Worker an implementation is presented, which offers simulation as a service in the cloud. The performance of the solution is illustrated by three application examples from the system design