Using Separability to Enable Time-Consistent Simulation of Distributed Space Systems

We consider the scheduling of parallel simulations of distributed systems based upon the notion of separability. We define a model of distributed systems that integrates a limited set of temporal properties, from which we deduce specific parallel simulation strategies that enable to compute the simulation schedule in advance, or even statically, and obtain an optimal rate of parallelism, while avoiding rollback and dynamic scheduling. This approach appears to be relevant for a large class of distributed computations, inasmuch as it relies upon a limited and rather usual set of temporal properties of the simulated systems.