Automated Optimization of Discrete Event Simulations without Knowing the Model

Modeling and simulation is an essential element in the research and development of new concepts and products in any domain. The demand for the development of more and more complex systems drives the complexity of the simulation models as well, which in turn urges the research of methodologies to reduce the execution times to a feasible amount. Two major types of resources can be exploited for acceleration: Multiple computing instances can be used to distribute the workload and perform independent computations simultaneously. Workload stemming from redundant computations can be avoided altogether by exploiting the presence of unused main memory to store intermediate results.

We observe that typically during development of simulation models neither time and resources nor required expertise is available to apply sophisticated optimization manually. We conclude that it is of utmost importance to investigate approaches to speed up simulation automatically. The most prevalent challenge of automating optimization is that, at the time of researching and developing the acceleration concepts and tools, the model is not yet available, and we need to assume that the model will not be implemented for a specific optimization technique. Hence, our methodologies need be devised without the model at hand, which can only be used by the finally implemented optimization tool at its runtime. We discuss how computer simulations can be automatically accelerated using either of the two optimization potentials mentioned above (multiple computing instances, available memory) without the model being provided at the time of researching the concepts and developing the tools. Finally, we discuss how by combining the two optimization vectors the full power of both can be unleashed at the same time.