A Strategy for Parallel Simulation of Declarative Object-Oriented Models of Generalized Physical Networks

For several years now, most of the growth of comput- ing power has been made possible by exploiting paral- lel CPUs on the same chip; unfortunately, state-of-the-art software tools for the simulation of declarative, object- oriented models still generate single-threaded simulation code, showing an increasingly disappointing performance. This paper presents a simple strategy for the efficient com- putation of the right-hand-side of the ordinary differen- tial equations resulting from the causalization of object- oriented models, which is often the computational bottle- neck of the executable simulation code. It is shown how this strategy can be particularly effective in the case of gen- eralized physical networks, i.e., system models built by the connection of components storing certain quantities and of components describing the flow of such quantities between them.