Distribution, Reuse and Interoperability of simulation models in heterogeneous distributed computing environments

Abstract

Modeling and Simulation (M&S) is gaining a central role in several industrial domains such as automotive, e-science and aerospace, due to the increasing complexity of system requirements and thus of the related engineering problems. Specifically, M&S methods, tools, and techniques can e↵ectively support the analysis and design of modern systems by enabling the evaluation and comparison of di↵erent design choices against requirements through virtual testing; this opportunity becomes even crucial when complete and actual tests are too expensive to be performed in terms of cost, time and other resources. Moreover, as systems result from the integration of components which are often designed and manufactured by di↵erent organizations belonging to di↵erent engineering domains (including mechanical, electrical, control, and software), great benefits can derive from the possibility to perform simulations which involve components independently developed and running on di↵erent and possibly geographically distributed machines. Indeed, distributed simulation promotes an e↵ective cooperative, integrated and concurrent approach to complex systems analysis and design. Although M&S o↵ers many advantages related to the possibility of doing controlled experiments on an artificial representation of a system, its practical use requires to face with important issues such as, (i) difficulties to reuse simulation models already made; (ii) lack of rules and procedures by which to make interoperable models created with di↵erent simulation environments; and, (iii) lack of mechanisms for executing simulation models in distributed and heterogeneous environments. Indeed, there are di↵erent simulation environments both commercial and noncommercial highly specialized that allow the design and implementation of simulation models in specific domains. However, a single simulation environment is not able to manage all the necessary aspects to model a system when it is composed of several components. Typically, the modeling and simulation of such systems, whose behavior cannot be straightforwardly defined, derived and easily analyzed starting from the behavior of their components, require to identify and face with some important research issues.