Sim-Diasca is a discrete-time simulation engine aiming for maximum concurrency, relying on a mode of operation that is both parallel and distributed.

More precisely, Sim-Diasca, which stands for Simulation of Discrete Systems of All Scales, aims at performing large-scale discrete-time simulations of all kinds of complex systems. This may include industrial systems (like smart metering infrastructures), social ones (ex: customer behaviour), full ecosystems (ex: cities), etc.

For that, the engine, which is generic (domain-agnostic), is both parallel and distributed, and focuses notably on scalability, in order to handle simulation cases which may be very large (potentially involving millions of interacting instances of models), while still preserving essential simulation properties, like causality, total reproducibility and some kind of ergodicity (to be understood here as the ability for the engine at each run to select in a reasonably fair manner one trajectory of the target system among the many it could experience).

The simulation engine is implemented in the Erlang language, and its execution platform of choice is GNU/Linux, from laptops to high-performance computing resources (ex: clusters).

It has been used in various projects, including the CLEVER project, and has been the main case study of the RELEASE project.

It is developed by EDF R&D and is released under the LGPL licence.

For more information, please refer to the Sim-Diasca General-Purpose Presentation, to the Sim-Diasca Technical Manual and to the Sim-Diasca Developer Guide (all of which are provided when requesting the sources of the engine, see below).

Contact us

Please contact us directly in order to get a source archive of the last stable release of Sim-Diasca and its associated documentation (just click below on the Contact us link, and fill in the corresponding form).